Distributed Computing, Artificial Intelligence, Bioinformatics, Soft Computing, and Ambient Assisted Living

Neuro-control which adopts neural network architectures to synthesis of control has been summarized and its application to electric vehicle control is developed in this paper. The neuro-control methods adopted here is based on proportional-plus-integral-plus-derivative (PID) control, which has been adopted to solve process control or intelligent control. In Japan about eighty four per cent of the process industries have used the PID control. Using the learning ability of the neural network, we will show the self- tuning PID control scheme (neuro-PID) and the real application to an electric vehicle control. environment.

MAS have already more than proved their effectiveness while dealing with high level distributed problems, but some domains (usually low level ones) are still reluctant to their use, usually on a performance basis. UAV’s multisensor integration systems take information coming from different sensors and integrate them into one global positioning solution, with a pre-analyzed fixed data fusion architecture topology in a changing environment. In this paper we will propose a novel adaptative MAS data fusion architecture for this problem, able to change its topology according to its conditions, and thus effectively improving the overall quality of the system.

This paper presents a work in progress about the design and development of a multiagent system for an autonomous vehicle (VERDINO). This vehicle (a standard golf cart) has been provided with many different sensors and actuators. The future multiagent system is intended to manage the data provided by the sensors and act on steering orientation and brake and throttle pedals.

This paper proposes a distributed multiagent architecture based on philosophical model of human practical reasoning for web search engine. The paper handles incompletely specified goals using

Belief Desire

and

Intention

model of practical reasoning in two stages. In the first stage deliberation cycle begins by accepting input query from the user and then presents the user with a set of semantically identified relevant topics obtained by exploring the surface and deep web. The chosen topic becomes the intention of the user. In the next stage, a set of partially ordered sequence of actions with respect to this intention are presented to the user using means-ends reasoning .This leads to incremental satisfaction of a user’s request via multiple iterations.

There is a general agreement in the fact that Agent Oriented Software Engineering (AOSE) needs development process definition for an accurate process management. The main trends in the field identify process and methodology in order to approach the process definition. This paper focusses in the idea that process and methodology must be considered independently. This means that not only the same process can be used for different methodologies but also that the same methodology can be used following different processes. The most suitable process can be selected by developers depending on several factors such as: human resources available, time restrictions, costs, etc. The previous approach is justified introducing a case study, which shows how different development processes can be applied while the team is following the same methodology (in particular, INGENIAS methodology).

In this paper a study of two approaches of a meta-algorithm, Meta_CHC_RBF, is presented. The main goal of this algorithm is to automatically design Radial Basis Function Networks (RBFNs) finding a suitable configuration of parameters (automatically adapted to every problem) necessary for the algorithm EvRBF, an evolutionary algorithm for the automatic design of asymmetric RBFNs. The principal difference between two proposals is the type of codification, in the fist one, the meta-algorithm uses binary codification, while in the second one, it implements real codification; affecting this influence of the codification kind in the carried out experimentation. Finally, results show that the first approach yields good marks reducing the computation time, with respect the second one.

This work presents a set of improvements and a performance analysis for a previously designed multi-objective optimisation parallel model. The model is a hybrid algorithm that combines a parallel island-based scheme with a hyperheuristic approach in order to grant more computational resources to those schemes that show a more promising behaviour. The main aim is to raise the level of generality at which most current evolutionary algorithms operate. This way, a wider range of problems can be tackled since the strengths of one algorithm can compensate for the weaknesses of another. A contribution-based hyperheuristic previously presented in the literature is compared with a novel hypervolume-based hyperheuristic. The computational results obtained for some tests available in the literature demonstrate the superiority of the hypervolume-based hyperheuristic when compared to the contribution-based hyperheuristic and to other standard parallel models.

New trends in parallel computing are moving towards multicore processors. A new factor then arises in such scenario,

concurrence

. But concurrence is not parallelism. Parallel applications that wish to take advantage of this new environment need to take this into consideration, or being completely rewritten in such a way that parallelism can be expressed by means of concurrence. In this evolving scenario, abstractions may help to keep performance. This paper shows how abstractions play an important role from the performance and scalability perspective when used to model the problem.

A mobile ad hoc network can be defined as a network that is spontaneously deployed and is independent of any static network. The network consist of mobile nodes with wireless interfaces and has an arbitrary dynamic topology. In this paper we present a toke- based group

k

-mutual exclusion algorithm for mobile ad hoc networks. The G

k

-ME problem is concerned with controlling the concurrent accesses of some resources by at most

k

nodes with the constraint that no two distinct resources can be accessed simultaneously. The proposed algorithm is adapted from the

RL

algorithm. The algorithm ensures the mutual exclusion, the bounded delay, and the

k

-concurrent entering property.

The W3C Recommendation for Semantic Annotations in WSDL and XML Schema (SAWSDL) defines an extension that can help to disambiguate the description of Web Services during automatic discovery and composition. In this way, SAWSDL is useful to facilitate the grounding stage when annotating Web Services. Despite SAWSDL does not specify a language to represent the semantic models for annotations, most of the times, ontologies are used to do it. In this work we propose a mechanism to automatically enrich SAWSDL annotations using concepts from different ontologies. As result, we provide a method for helping experts to annotate web services according to the SAWSDL recommendation.

In P2P systems, nodes typically connect to a small set of random peers to query them, and they propagate those queries along their own connections. To improve that mechanism, Semantic Overlay Networks influence those connections depending on the content of the peers, clustering peers in overlapped groups (Semantic Overlay Networks). Ontologies are used for describing semantic information of shared items and user profile. Once the peers are grouped by their semantic information, we can take advantage of that distribution to add some new functionalities as recommendation. In this paper we focus on the description and evaluation of SONAtA, a SON classifier algorithm to globally organize peers into semantic groups, executed locally in each peer.

In a Multi-Agent system, middleware is one of the components used to isolate control and communications. The use of standards in the implementation of an intelligent distributed system is always advantageous. This paper presents a middleware that provides support to a multi-agent system. Middleware is based on the standard Data Distribution Services (DDS), proposed by Object Management Group (OGM). Middleware organizes information by tree based ontology and provides a set of quality of service policies that agents can use to increase efficiency. DDS provides a set of quality of service policy. Joining quality of service policy and the ontology allows getting many advantages, among others the possibility of to conceal some details of the communications system to agents, the correct location of the agents in the distributed system, or the monitoring agents in terms of quality of service. For modeling the middleware architecture it has used UML class diagrams. As an example it has presented the implementation of a mobile robot navigation system through agents that model behaviors.

The ontology evaluation utilities that are currently available allow the user to check the internal consistency of an ontology, its syntactical correctness and, at most, the fulfillment of some philosophical constraints related to rigidity or identity. However, there is no contribution in the ontology evaluation field that proposes a method to dynamically test ontologies with regard to their functional specification. Thus, no software for this task has been built until now. This paper presents a tool, OntologyTest, designed to overcome this drawback. The tool allows the user to define a set of tests to check the ontology’s functional requirements, to execute them, and to inspect the results of the execution. The whole set of tests (or a particular test) can be executed at any time; thus it simplifies the testing of ontology both during its development and during its evolution.

SystemC is a library that facilitates the development of Transaction Level Models (TLM). These models are composed of both hardware and software components. This library allows designing and verifying hardware system components at a high level of abstraction. This supports the development of complex systems. A real industry SystemC model usually contains a high number of functional blocks which increase its simulation run time. SystemC executes only one process at any time, even if the hardware supports execution of concurrent processes. In this paper we present a new methodology for distribution of the simulation of complex models in a parallel computing system. We apply our own approach in a real industry SystemC model of a Power Line Communication (PLC) network.

Snapshot algorithms are fundamental algorithms in distributed computing. However, most existing snapshot algorithms are designed for a static network system in which the set of channels and the processes at the endpoints of the channels do not change with time. They cannot be applied directly in recent emerging MANETs, which usually have no fixed infrastructure and may experience dynamic topology changes in the mid of execution. In this paper, we report recent results on developing snapshot algorithms for MANETs.

This paper presents SYLPH, a novel distributed architecture which integrates a service-oriented approach into Wireless Sensor Networks. One of the characteristics of SYLPH is that it can be executed over multiple wireless devices independently of their microcontroller or the programming language they use. SYLPH works in a distributed way so that most of the application code does not have to reside in a central node. Furthermore, SYLPH allows the interconnection of several networks from different wireless technologies, such as ZigBee or Bluetooth. This paper focuses on describing the main components of SYLPH and the issues that lead to design and develop this new approach. Results and conclusions are presented after evaluating a preliminary version of this architecture.

Parking behavior is an interesting field of study. Councils are concerned about encouraging a responsible use of the parking space, which, due to our habits in modern life, is more and more solicited. Companies with exploitation rights always seek how to get more profit from the parking lots that they control. Our system is designed to accomplish two goals: measuring parking behavior in terms of parking space occupation and vehicle rotation in specific areas of parking, and estimating possible incomes from the exploitation of a parking lot.

This work reveals the benefits obtained from combining common-sense reasoning and multi-agent systems on top of a fully equipped middleware platform. The architecture here proposed is founded on the service composition paradigm, as the comprehensive solution to relieve users from being involved in system decision making. In this regard, the environment and domain understanding is emulated by the common-sense reasoning engine that supports the multi-agent system on the task of effectively accomplishing the actions that fullil the new arisen requirements.

This paper presents a systematic literature review to investigate which techniques have been applied to give support to the requirements engineering activity in the development of Multi-Agent Systems (MAS). We reviewed 49 of 389 papers found that were directly related to our goal. The results show that most of the proposals for dealing with requirements (78%) use already defined methods or techniques, and that 67% of these techniques are based on the goal-oriented paradigm. A total of 96% of the reviewed papers focus on techniques for analyzing requirements and only 39% of them explicitly consider some kind of elicitation technique. The results are important for determining current research activities in Requirements Engineering for MAS and for the identification of research gaps for further investigation.

Testing is an important activity in software development in order to assure the correctness of software. However, testing is often disregarded in most agent oriented methodologies, mainly because they focus on analysis and design activities, and consider that implementation and testing issues can be performed using traditional techniques. But multi-agent systems implementation has some features that make it distinctive from traditional software. This paper presents an overview of testing in agent orientation based on the V-Model in order to establish the role of testing activities in an agent oriented development lifecycle. It also identifies how different types of testing are covered by previous work and the directions for further work.

Components in open environments are reusable and loosely coupled, enter and leave organizations, but even the most reusable service is not useful if it cannot be found by those responsible for creating potential consumers. In this paper a Service Facilator Agent (SFA) is presented to deal with service management in Open Multi-Agent Systems. The SFA functionality is based on semantic web services and provides an accurate service composition based on planning techniques and being aware of service execution time.

A guideline for designing Virtual Organizations, composed of a set of phases for requirement analysis, structure design and organizational dynamics design, is presented in this paper. It has been applied to a personalized information system, viewed as an open multi-agent system that acts as a regulated information meeting-point.

This paper presents a Java framework to implement distributed applications via Bluetooth. It provides a high-level Application Programming Interface (API) which simplifies the creation of applications for Bluetooth devices in Java ME and Java SE platforms. This framework is based in a client-server architecture and an event-driven asynchronous communication mechanism. As an example of use, we solve two well-known evolutionary computation problems (the Traveler Salesman Problem and the Wave Function Problem).

Symbolic Regression is one of the most important applications of Genetic Programming, but these applications suffer from one of the key issues in Genetic Programming, namely bloat – the uncontrolled growth of ineffective code segments, which do not contribute to the value of the function evolved, but complicate the evolutionary proces, and at minimum greatly increase the cost of evaluation. For a variety of reasons, reliable techniques to remove bloat are highly desirable – to simplify the solutions generated at the end of runs, so that there is some chance of understanding them, to permit systematic study of the evolution of the effective core of the genotype, or even to perform simplification of expressions during the course of a run.

This paper introduces an alternative approach, Equivalent Decision Simplification, in which subtrees are evaluated over the set of regression points; if the subtrees evaluate to the same values as known simple subtrees, they are replaced. The effectiveness of the proposed method is confirmed by computer simulation taking simple Symbolic Regression problems as examples.

The field of communication networks has witnessed tremendous growth in recent years resulting in a large variety of combinatorial optimization problems in the design and in the management of communication networks. One of these problems is the terminal assignment problem. The task here is to assign a given set of terminals to a given set of concentrators. In this paper, we propose a Hybrid Differential Evolution Algorithm to solve the terminal assignment problem. We compare our results with the results obtained by the classical Genetic Algorithm and the Tabu Search Algorithm, widely used in literature.

Microarray technology has provided biologists with the ability to measure the expression levels of thousands of genes in a single experiment. One of the urgent issues in the use of microarray data is the selection of a smaller subset of genes from the thousands of genes in the data that contributes to a disease. This selection process is difficult due to many irrelevant genes, noisy genes, and the availability of the small number of samples compared to the huge number of genes (higher-dimensional data). In this study, we propose an iterative method based on hybrid genetic algorithms to select a near-optimal (smaller) subset of informative genes in classification of the microarray data. The experimental results show that our proposed method is capable in selecting the near-optimal subset to obtain better classification accuracies than other related previous works as well as four methods experimented in this work. Additionally, a list of informative genes in the best gene subsets is also presented for biological usage.

In recent years, Machine Learning (ML) has witnessed a great increase of storage capacity of computer systems and an enormous growth of available information to work with thanks to the WWW. This has raised an opportunity for new real life applications of ML methods and also new cutting-edge ML challenges like: tackle with massive databases, Distributed Learning and Privacy-preserving Classification. In this paper a new method capable of dealing with this three problems is presented. The method is based on Artificial Neural Networks with incremental learning and Genetic Algorithms. As supported by the experimental results, this method is able to fastly obtain an accurate model based on the information of distributed databases without exchanging any data during the training process, without degrading its classification accuracy when compared with other non-distributed classical ML methods. This makes the proposed method very efficient and adequate for Privacy-Preserving Learning applications.

For the automatic inspection for printed labels, which are covered with rubber-like coatings and curl, we have developed a camera-based portable inspection system. In this paper, we explained the developed system, and especially discuss the inspection method of the spread and chip of the printed labels using neural networks. The experimental results confirm the validity of the proposed method for the spread and chip of alphanumerics.

In this paper, we present results ensuring the correct compositionality of the components (named capsules in UML-RT terminology) of a real-time system at different specification stages using the methodological framework called MEDISTAM-RT, which guarantees the temporal consistency and safe (deadlock free) communication between capsules. This allows the compositional verification of systems designed with this methodology, in such a way that the verification of a complete real–time system can be reduced to the verification of its smallest capsules.

In this work, a parallel version of the evolutionary algorithm called UEGO (

Universal Evolutionary Global Optimizer

) has been implemented and evaluated on shared memory architectures. It is based on a threaded programming model, which is suitable to be run on current personal computers with multicore processors.

The theoretic and algorithmic description of the parallel batch pattern back propagation (BP) training algorithm of multilayer perceptron is presented in this paper. The efficiency research of the developed parallel algorithm is fulfilled at progressive increasing of the dimension of parallelized problem on general-purpose parallel computer NEC TX-7.

Electron tomography allows structure determination of complex biological specimens. The tomographic reconstruction algorithms require an extensive use of computational resources and considerable processing time to compute high resolution 3D reconstructions. High performance computing (HPC) turns out to be essential to cope with these demands. We propose and evaluate different HPC strategies based on the well-known master/slave paradigm for tomographic reconstruction. Our results demonstrate that there is an underlying problem to tackle, if the performance is to be further improved: the access to the shared file system. On the other hand, it has been shown that it is possible to find out the optimal size of the tasks distributed by the master, specially for large datasets.

Parallel computing has become an important research field in the last years. The availability of hardware and the success of grid computing have motivated this interest. In this paper we present a new approach for managing parallel environments in a grid-like manner using agent technologies. Multi-agent systems provide an added value to standard grid approaches due to their high level abstraction and flexibility, which are exploited by our approach. As a result, a general purpose platform for parallel execution of tasks using mobile agents is introduced. A particular application of the platform for implementing a complex knowledge acquisition method is also introduced, and the computational benefits of the parallelisation are measured.

VS (Virtual Supervisor) Diagrams, defined from the FPM (Finite Positions Machines) framework, are used to model, analyze and validate automated manufacturing systems and they are obtained, in a practical way, from the PLC (Programmable Logic Controller) signals. This current paper presents a neural network architecture in order to identify that type of diagrams. It is made up of a supervised Hebb neural network cascade linked to a recurrent Hopfield network.

Self-organising neural networks preserves the topology of an input space by using their competitive learning. In this work we use a kind of self-organising network, the Growing Neural Gas, to represent non rigid objects as a result of an adaptive process by a topology-preserving graph that constitutes an induced Delaunay triangulation of their shapes. The neural network is used to build a system able to track image features in video image sequences. The system automatically keeps correspondence of features among frames in the sequence using its own structure.

The objective of the presented project was to develop and implement a forecasting instrument to predict the oil price in short-, mid- and long-term. Because there are a lot of different and complex factors influencing the oil price, the neural net method was chosen. Many data that could be relevant for the prediction was integrated in the net and several architecture models were tested. The data base consisted of about 2000 data records reflecting the period between 1999 until 2006. As result of the project it can be summarized that the implemented neural nets could not achieve sufficient results in the short-term forecasting but achieved very good results in the mid- and long-term predictions. Therefore it should be a valuable instrument for supporting management decisions in this field.

The Trace transform, a generalization of the Radon transform, allows one to construct image features that are invariant to a chosen group of image transformations. It consists of tracing an image with straight lines along which certain functionals of the image function are calculated. It can be useful to construct invariant features to rotation, translation and scaling of the image. In this paper, we demonstrate the usefulness of the features in classification of Jawi character images using multilayer perceptron neural networks. We compare the result of character recognition with those obtained by using features based on affine moment invariants.

In this paper we present an ensemble based translator which combines a simple rule and the connectionist translator RECONTRA. One of the problems of RECONTRA is the growing size of the networks as the tasks to be translated increase in size and complexity. A possibility to reduce this problem and increase the results is the use of ensembles. A simple rule (the presence of the symbol ‘¿’ in the input sentence) which allows separating the task in two parts that can be approached separately and, during the test phase allows integrating the results is employed.

We have developed an awaking behavior detection system using a neural network (abbreviated as NN). However, the detection ability of unknown people is not sufficient with compared to that of learned people. In this research, to improve the detection ability of unknown people, we apply an online tuning system using a continuous learning of the NN for the detection system. In the online tuning system, only a few additional data of a new objective person are used for the continuous learning, where the weights of the NN converged in the initial learning are used as the initial weights for the continuous learning. In this paper, to verify an ability of the online tuning system, we compare detection ability of the converged initial learning with that of the converged online tuning.

Many macroscopic properties: hardness, corrosion, catalytic activity, etc. are directly related to the surface structure, that is, to the position and chemical identity of the outermost atoms of the material. Current experimental techniques for its determination produce a “signature” from which the structure must be inferred by solving an inverse problem: a solution is proposed, its corresponding signature computed and then compared to the experiment. This is a challenging optimization problem where the search space and the number of local minima grows exponentially with the number of atoms, hence its solution cannot be achieved for arbitrarily large structures. Nowadays, it is solved by using a mixture of human knowledge and local search techniques: an expert proposes a solution that is refined using a local minimizer. If the outcome does not fit the experiment, a new solution must be proposed again. Solving a small surface can take from days to weeks of this trial and error method. Here we describe our ongoing work in its solution. We use an hybrid algorithm that mixes evolutionary techniques with trusted region methods and reuses knowledge gained during the execution to avoid repeated search of structures. Its parallelization produces good results even when not requiring the gathering of the full population, hence it can be used in loosely coupled environments such as grids. With this algorithm, the solution of test cases that previously took weeks of expert time can be automatically solved in a day or two of uniprocessor time.

The GPS System is not valid for positioning under a roof; therefore Wi-Fi positioning systems which allow the localization of a device inside a building are designed. The use of fuzzy logic is argued because there hasn’t been found any positioning system based in this technology, therefore we pretend to observe how its use in this fields works.

This article highlights the peak of geographic information systems (GIS). These systems use has grown these last years applying to a great number of sectors. In this document we will focus in the use of theses systems in the epidemiologic filed, where the use of this kind of technology is essential to identify and avoid possible threats.

Models are becoming first-class artifacts in Software Engineering because they provide better productivity and quality. In this paper we present a framework for developing all kinds of applications, mainly by following the best practices of the two main approaches to Model-Driven Engineering (MDE). On one hand is MDA (Model-Driven Architecture), proposed by the OMG (Object Management Group) and on the other hand are the Software Factories, proposed by Microsoft. Both approaches have their pros and cons and that is why we want to mix them into a single framework by selecting the best that each of them can give us. TALISMAN MDE Framework is based on the opinion of recognized experts in MDE and on the lessons learned in our previous work, TALISMAN MDA.

This paper is concerned with a new construction of an electronic nose system based on a neural network. The neural network used here is a competitive neural network by the learning vector quantization (LVQ). Various smells are measured with an array of many metal oxide gas sensors. After reducing noises from the smell data which are measured under the different concentrations, we take the maximum values among the time series data of smells. The data are affected by concentration levels, we use a normalization method to reduce the fluctuation of the data due to the concentration levels. Those data are used to classify the various smells of tees and coffees. The classification results are about 96% in case of four kinds of tees and about 89% for five kinds of coffees.

Models and transformations between models are provided as the core of Model-Driven Engineering, offering reusability of knowledge and processes. In order to establish the basis of future advances in this emerging paradigm, this paper is focused on the principles of meta-models and transformation models. Moreover, the concept of meta-model is becoming an essential artifact for MDE based solutions, thus we have centered our background review in the state of art related to meta-model specifications and model transformation technologies. Our research is aimed at getting a higher degree of interoperability among available meta-model specifications by raising the transformation models to the upper meta-layers. Some conclusions extracted suggest that this is still an early solution which demands greater efforts in terms of research, development and specification, with many interesting open subjects like design of generic editors for model-agnostic visual modeling, integration of model instances from different meta-models, improvements of the semantic knowledge offered by present modeling languages or even the evaluation of the applicability of graph transformation techniques towards formal transformation models.

Driver development is a tedious and complex task, which involves deep knowledge of the operating system and the programming language needed to communicate with hardware devices. Due to the vertiginous advances inside the hardware industry, it is very important to develop drivers in an easy and fast way. But the reality shows that when we develop a driver there is little information available and, what is worse, that information is wrong or inaccurate. Thus, trying to implement hardware programs became almost impossible.

Both Information Compression and Parallel Processing are key issues nowadays. And both are our framework for our proposal. Our purpose was to accomplish an exhaustive study about patterns recognition using backpropagation neural networks with multilayer perceptron topology for its subsequent application to image and video compression. The mathematical study of learning algorithms and construction of neural nets encompassed the greater part of our time for this study. The next step was to design the most appropriate neural net for our objective through the paradigm of the oriented-object programming.

In this paper, a distributed intelligent monitoring system for a micro-scale turning process is presented. A fuzzy model, running on a distributed architecture, helps on decision-making about the imbalance degree of the spindle in the ultra-precision diamond turning process. The suitability of a three-input/single-output fuzzy model is assessed directly on an ultra-precision lathe, verifying it effectiveness. A brief explanation of the single point diamond turning process and problems caused by the vibrations generated due to the imbalance of the spindle, are also presented. Real-time application of the fuzzy decision making, as part of a networked distributed monitoring system, assists the operator when manufacturing complex parts.

There are two main contributions of this paper. The first one is a particular approach to adapting software agents and graph theory to the supply chain configuration in personal computer industry. The second contribution is the simulation of the DyConSC model in the NetLogo platform as well as the presentation of dependencies and conclusions.

The paper presents a concept of an IT tool for assessing manufacturing processes in automotive industry, which is based on a sequence block quality indicator. The background of researches is presented and distributed IT environment in context of Just-in-Sequence concept is described. The concept of sequence block quality and calculating solution are explained. IT application in distributed manufacturing environment is presented. Finally conclusions and future works are stated.

The aim of this paper is to discussed the potential of the new and emerging solutions that might be successfully applied for improvement of manufacturing system management in highly volatile environment of today’s global economy. In order to remain competitive in the market manufacturing companies have to provide a quick-response to crisis economy shifts. Authors discuss the conditions that have to be met by new standards of IT systems supporting the operations management. The main characteristics of new IT solutions are defined.

In this paper we discuss the idea of Web Service market places as an extended point of view towards the SOA paradigm. As existing business applications include software functionality which can be reused in new applications the idea of a shared management among allied (federated) network nodes as participants of a superordinated union of software providers and software consumers arises. Having a look at the example of a Federated ERP system we show the advantages and challenges coming along with this idea. As one of the most important issues to be considered in this context we discuss a decentralized approach for the establishment of trust relationships in federated networks. In this approach Web Service consumers rate Web Service providers, that support the competitive behavior between Web Service providers in a market-oriented system.

GRAIXPERT is a hybrid expert system of the GRAI Methodology (used for enterprise modelling) based on different reasoning (Case-based reasoning (CBR), decomposition, transformation) containing rules, reference models and a case base. It enables to aid specialists of enterprise modelling during a study. This paper presents the concepts of this tool. We focus our attention on the different reasoning used for this tool. Other tools (GRAISUC and GRAIQUAL) are also being developed for completing the expert system The objective is to be efficient for the improvements of the performance (quality, cost and lead time) of the enterprise in terms of diagnosis and design. Finally we complete the presentation with an example in order to illustrate the concepts presented.

The present customer consumption models with short product’s life cycle result in increasing number of used products that need to be collected and reused or disposed. In order to gain the advantages of “two-way” economy a change in business approach is needed. Many closed-loop supply chains lack good product and inventory data, which causes inefficiencies. Therefore specific tools are necessary that allow efficient management of information accompanying material flows. In this paper author briefly presents the theoretical background of the research. The common problems that appear by closing supply loop are described. The potential for application of agent-based solutions in order to improve the reverse logistics management is discussed. The concept of agent-based solutions is presented.

The paper proposes a set of tools to be used to deal with complexity of resources and capacity management problem. The set includes scenarios developed by experts to analyse possible future market conditions and customers requirements, as well as agent-based modelling and simulation used to show and analyse dynamic relation between company’s resources and those between the resources and environment company is functioning in. The techniques presented are supposed to benefit from distributed knowledge of the experts employed to develop scenarios and of capabilities of software agents, including those of adapting dynamically to changing business environment.

This paper presents a web-based system of membership registration for the Japan Volleyball Association (JVA). There are about 0.3 million JVA members, comprised of children, student, housewife, and professional players, along with staff and others. Furthermore, the members of teams belonging to JVA number about 30,000, and there are about 500 groups divided by district or age group. This web-based system achieved unified management of member information and competition administration support for such a large, complicated organization.

In this paper, we present a web application development framework that can improve efficiency of web application user interface (UI) development less affected by the change of web technologies. Recently, RIA (Rich Internet Applications) has gotten a lot of attention. Dynamic UI is essential of RIA, but RIA programming is complicated and varies depending on its implementation. To employ such a new technology, how to reduce the cost increase and the efficiency degradation is important. We propose MVC(Model-View-Controller)-based UI model that can represent not only static UI but also dynamic UI independently from specific RIA implementation. In the upper process of development, developers can design UI with UI model, so they don’t have to learn detail of RIA. In the lower process, a code generator generates specific RIA implementation code from UI model. Therefore programmers just have to write a code for customization. Evaluation result for sample application development shows programmers have only to program 25% of total lines of code for client-side.

In recent years, various enterprise information systems utilizing internet have been developed. Nevertheless legacy systems also have been still running in enterprise. The data combination between both systems has become to be important. There are some considerable points in practically system developing. This paper investigates several methods and proposes the data replication method to develop the high-quality and efficient combination system based on the experience of real system developments.

This paper is a survey report about why students use online video services. Based on the Technology Acceptance Model (TAM), a hypothesis model for online video services users was designed. The hypothesis model incorporates social influence, flow experience, comfortable communication, and advertisement interference into the basic TAM. With a questionnaire given to about 350 students, behaviors of online video service users were analyzed.

Ant Colony Optimization is a metaheuristic which has been successfully applied to solve several NP-hard problems. It includes several algorithms which imitate the behavior of natural ants. The algorithm called Ant Colony System is one of the best-performing ant-based algorithms. In this paper we present an enhanced algorithm, which applies dynamic programming to improve the solution generated by the ants. The method is applied to the well-known Traveling Salesman Problem. We present computational results that show the improvement obtained with the modified algorithm.

Quality of service policies in communications is one of the current trends in distributed systems based on middleware technology. To implement the QoS policies it is necessary to define some common parameters. The aim of the QoS policies is to optimize the user defined QoS parameters. This article describes how to obtain the common QoS parameters using message queues for the communications and control components of communication. The paper introduces the “Queue-based Quality of Service Cycle” concept for each middleware component. The QoS parameters are obtained directly from the queue parameters, and Quality of Service Policies controls directly the message queues to obtain the user-defined parameters values.

Planning is a difficult computational problem. One way to increase efficiency of searching for a solution may be a transformation of a problem to another problem and then search for a solution of the transformed problem. In this work a transformation of STRIPS planning problem with uncertainty of operators outcomes to linear programming is shown. The transformation from planning to Linear Programming is based on mapping of conditions and operators in each plan step to variables. Exemplary simulation shows properties of proposed approach.

DNA computing is a new computing paradigm which uses bio-molecules as information storage media and biochemical tools as information processing operators. This field has shown many successful and promising results for various applications. Since DNA reactions are probabilistic in nature, different result could be produced even in the same situations, which can be regarded as errors in computing. In order to overcome the drawbacks, many works have focused on the design or error-minimized DNA sequence to improve the reliability of DNA computing. Although the design of DNA sequences is dependent on the protocol of biological experiments, it is highly required to establish a method for the systematic design of DNA sequences, which could be applied to various design constraints. In the previous paper, Ant System approach has been proposed to solve the DNA sequence optimization problem. In this paper, the optimized parameters of Ant System approach are searched to improve the performance of the Ant System for DNA sequence optimization.

In DNA based computation and DNA nanotechnology, the design of good DNA sequences has turned out to be an essential problem and one of the most practical and important research topics. Basically, the DNA sequence design problem is a multi-objective problem, and it can be evaluated using four objective functions, namely,

H

measure

, similarity, continuity,

and

hairpin

. There are several ways to solve a multi-objective problem, such as value function method, weighted sum method, and using evolutionary algorithms. However, in this paper, common method has been used, namely weighted sum method to convert DNA sequence design problem into single objective problem. Binary particle swarm optimization (BinPSO) is proposed to minimize the objective in the problem, subjected to two constraints:

melting temperature

and

GC

content

.

Based on experiments and researches done, 20 particles are used in the implementation of the optimization process, where the average values and the standard deviation for 100 runs are shown along with comparison to other existing methods. The results obtained verified that BinPSO can suitably solve DNA sequence design problem using the proposed method and model, comparatively better than other approaches.

In this paper we present a model to distributed feature selection problem (DFSP) based on ACO and RST. The algorithm looks for reducts by using a multi-colony ACO as search method and RST offers the heuristic function to measure the quality of one feature subset. General results of using this approach are shown and formers results of apply ACO and RST to the feature selection problem are referenced.

In the last years, the market is demanding (scientific, multimedia, real-time, etc.) applications with high bandwidth requirements. To support this, the bandwidth of the network links has increased to reach multiple gigabit per second. Nevertheless, taking advantage of multigigabit per second links requires a lot of processor cycles for communication tasks, diminishing the processor cyles that remains available for the application. Actual multiprocessor and multicore architectures as well as programmable NICs (Network Interface Cards) provide new opportunities to exploit the parallelism, distributing the communication overhead across the processors available in the node. We have designed a network interface that takes advantage of the different processors available in the node. In this paper, the advantages of this optimized network interface are shown by analyzing the performance of a web server application.

Agrega is a SCORM 2004 learning object repository network with the LOM-ES profile, with nodes distributed throughout all of Spain’s Autonomous Regions. Each repository offers a set of services for managing the objects it stores. From the point of view of interoperability, the project had two basic objectives: one was to create a modularised functionality that could be reused by other external applications and the other was to be able to interoperate with other external digital repositories. The first objective has been met by implementing a web service architecture used by the nodes and any external application that wishes to integrate with Agrega; in order to meet the second objective, it has been decided to implement IMS DRI and SQI standards, which guarantee interoperability between repositories. This article describes the infrastructure for web service interoperability and standards that have been implemented.

The emergence of new technologies provides the opportunity to develop novel solutions that facilitate the integration of the visual disabled people in different activities of our daily life. This paper presents a distributed intelligent architecture, called DIAMI, focused on facilitating the integration of blind musicians in orchestras. The DIAMI architecture provides a distributed, ubiquitous system aimed at providing a way for blind musicians to receive the instructions of the orchestra conductor in an unobstructive manner. The structure of the DIAMI architecture and the preliminary results obtained are presented in detail within this paper.

The objective of this paper is to present an improvement of a decision-making support system based in inductive learning, applied to risk prevention in maritime works. The improvement shown here is based on the redefinition of training examples structured as temporal windows over certain attribute values.

Weather forecasting [12] has been one of the most scientifically and technologically challenging problems around the world in the last century. This is due mainly to two factors: firstly, the great value of forecasting for many human activities; secondly, due to the opportunism created by the various technological advances that are directly related to this concrete research field, like the evolution of computation and the improvement in measurement systems. This paper describes several techniques belonging to the paradigm of artificial intelligence which try to make a short-term forecast of rainfalls (24 hours) over very spatially localized regions. The objective is to compare four different data-mining [1] methods for making a rainfall forecast [7], [10] for the next day using the data from a single weather station measurement.

We propose in this paper the design and implementation of a Data Grid and its deployment in the Retelab project. Retelab Data Grid was deployed as a portlet in the Retelab web portal, adding new functionalities to the project. Such a system allows the users to search and manage the geoscientific data stored in the Retelab datasets through data attributes. Thus, the users do not need to know where or how data are stored. Retelab users can upload their own data and label them with metadata, which is a way of sharing information through the Ocean research community. We developed some procedures to use data as parameters in running jobs in the Grid system using a comfortable and visual interface. The Data Grid also provides different means for analyzing and visualizing data stored in Retelab.

The application of microarray data for cancer classification has recently gained in popularity. The main problem that needs to be addressed is the selection of a smaller subset of genes from the thousands of genes in the data that contributes to a disease. This selection process is difficult because of the availability of the small number of samples compared to the huge number of genes, many irrelevant genes, and noisy genes. Therefore, this paper proposes an improved binary particle swarm optimisation to select a near-optimal (smaller) subset of informative genes that is relevant for cancer classification. Experimental results show that the performance of the proposed method is superior to a standard version of particle swarm optimisation and other related previous works in terms of classification accuracy and the number of selected genes.

In this paper a new mathematical model to simulate the computer virus spreading on a network is introduced. It based on cellular automata on graphs. Specifically the model proposed is a SEIS model where each node/computer of the network can be in one of three states: susceptible, exposed and infected.

Ant-based algorithms constitute a metaheuristic successfully applied to solve combinatorial optimization problems. This paper presents the computational results of the application of an ants-algorithm, called the Rank-Based Ant System, to solve the Undirected Rural Postman Problem. Because this is a NP-hard problem, several approximate methods have been proposed to solve it. The results reported in this paper improve some of the ones reached by other approximate methods applied to the problem.

Computer security has become a major problem in our society. In particular, computer network security is concerned with preventing the intrusion of an unauthorized person into a network of computers. An intrusion detection system (IDS) is a tool to monitor the network traffic and users’ activity with the aim of distinguishing between hostile and non-hostile traffic. Snort is an IDS available under GPL, which allows pattern search. This paper presents a new anomaly pre-processor that extends the functionality of Snort IDS, making it a hybrid IDS.

In multipoint real-time communication, one member needs to transmit information to all of the other members. However, there are some problems in real-time multicasting. Because the network resource of each member is different among multicast members in heterogeneous access environment, the maximum quality of the video stream, which each member can receive, is also different among them. Thus, we propose a new algorithm to fairly determine the rate of each layer according to network resources of each member in the shared-tree type application-level layered multicast. We aim to guarantee the maximal quality of service of each member’s communication according to each network resource. In addition, it is confirmed that our proposed method is effective through the quantitative performance evaluation.

Due to the increasing fuel price, the European fishing sector has been suffering a descendent trend since 1998. It is essential to find efficient solutions, through R&D, by the application of new technologies. This paper presents a portable, scalable and reusable data acquisition system for the categorization of energy consumption distribution in fishing vessels. Furthermore tools for processing, displaying and spreading the collected data have been developed. The resulting information will enable further analysis in order to draw energy savings and energy efficiency improvements.

Clinical Practice Guidelines have been designed to reduce uncertainty in the medical decision making in order to improve medical care and reduce costs. A way to facilitate and formalize this task is by translating it into the computers domain. This paper introduces a validator for clinicial practice guidelines as part of a Decision Support System. So, the clinical staff can introduce patient data for a particular disease, and the system is able to validate and find inconsistencies in the application of Clinical Practice Guideline for such disease.

In this paper, we present an adaptation of Gaussian Processes for learning a joint probabilistic distribution using Bayesian Programming. More specifically, a robot navigation problem will be showed as a case of study. In addition, Gaussian Processes will be compared with one of the most popular techniques for machine learning: Neural Networks. Finally, we will discuss about the accuracy of these methods and will conclude proposing some future lines for this research.

Microshrinkages are known as probably the most difficult defects to avoid in high-precision foundry. The presence of this failure renders the casting invalid, with the subsequent cost increment. Modelling the foundry process as an expert knowledge cloud allows properly-trained machine learning algorithms to foresee the value of a certain variable, in this case the probability that a microshrinkage appears within a casting. Extending previous research that presented outstanding results with a Bayesian-network-based approach, we have adapted and tested an artificial neural network and the K-nearest neighbour algorithm for the same objective. Finally, we compare the obtained results and show that Bayesian networks are more suitable than the rest of the counterparts for the prediction of microshrinkages.

The advent of the Internet has opened a scope for research in new methods and tools that may facilitate the teaching and learning processes. This has, in turn, led to the development of learning platforms to support teaching and learning activities. Nowadays most universities provide their academic community with some form of a learning management system (LMS). To achieve the optimal use of such type of systems, they must integrate all their academic community and preexisting applications at its institutions. These complex objectives can be reached by using a robust architecture, preferably an open system, based on distributed computing. In this paper, we expose the Universitat de València particular case: The integrated LMS implementation, architecture base, and hardware four years evolution through distributed computing to better adequate to community requirements.

In this work it is addressed the problem of information evaluation and decision making process in Group Decision Support Systems (GDSS). A Multi-valued Extended Logic Programming language is used for imperfect information representation and reasoning. A model embodying the quality evaluation of the information, along the several stages of the decision making process, is presented. This way we give the decision makers a measure of the value of the information that supports the decision itself. This model is presented in the context of a GDSS for VirtualECare, a system aimed at sustaining online healthcare services. Reasoning with incomplete and uncertain knowledge has to be dealt with in this kind of environment, due to the particular nature of the healthcare services, where the awful consequences of bad decisions, or lack of timely ones, demand for a responsible answer.

An approach to improving the management of complexity during the redesign of technical processes by means of MBR and CBR techniques is proposed. The key point of this approach is the multi-model hierarchical representation to group items of equipment of a technical process in functional sections according to their functions and intentions. A CBR system gives similar equipment/functional sections to the one selected by the user to be replaced/modified. The output is a set of candidate equipment/functional sections to be adapted into the original process.

This paper presents a set of software tools called that helps in the building knowledge in an autonomous meaningful learning process through an open student model and a student conceptual map explorer. This tool uses adaptive tests based on a Progressive Inquiry (PI) model and has been used for teaching Computer Architecture (in particular the domain of Computer memory hierarchy) although the proposed system is supposed to be valid for any knowledge domain. The mentioned field has been chosen due to its limited complexity but it is sufficiently general to be learnt in several disciplines with different levels of detail.

The simulation of the main molecular operations used in DNA Computing can lead the researchers to develop complex algorithms and methods without the need of working with real DNA strands in-vitro. The purpose of this paper is to present a computer program which simulates an electrophoresis process over DNA molecules which is an essential operation for the identification of DNA strands. This simulation represents a useful tool for a virtual laboratory which is oriented to DNA computations. A wide variety of variables are taking into account like voltage, friction, temperature and viscosity of the gel used. The results given by the software can show the behavior of a DNA electrophoresis under certain physical conditions which allow us to obtain the relative size of the molecules involved and the best parameters to carry out the operation in-vitro efficiently.

The bills in circulation generate a large amount of fatigue bills every year, causing various types of problems such as the paper jam in automatic tellers due to overworked and exhausted ones. An advanced technique is requested in order to classify the levels of fatigue as well as distinguish between the used and the new ones. Therefore, the purpose of this paper is to present the classification method of fatigue bills based on support vector machine(SVM) by using acoustic signals. The effectiveness of this approach is demonstrated by the bill identify experimentation based on the real acoustic signal.

Recycling packaging waste requires the collection of the waste in an efficient way. We can either dispose of our packaging waste at local packaging collection places or use the recycling bins that are provided along the streets in some municipalities. These bins are emptied and cleaned out periodically by the municipality workers. This paper describes the application of artificial ants to define optimal paths to collect the bins. We consider the particular case of waste recycling in the province of Zamora, and we apply the method to a city called Benavente.

Firearm identification is one of the most essential, intricate and demanding tasks in crime investigation. Every firearm, regardless of its size, make and model, has its own unique ‘fingerprint’ with respect to the marks on fired bullet and cartridge cases. In this study, we investigate the features extracted from the images of the centre of the cartridge case in which firing pin impression is located. Geometric moments up to the sixth order were computed to obtain the features based on a total of 747 cartridges case images from five different pistols of the same model. These sixteen features were found to be significantly different using the MANOVA test. Correlation analysis was used to reduce the dimensionality of the features into only six features. Classification results using cross-validation show that about 74.0% of the images were correctly classified and this demonstrates the potential of using moment based features for firearm identification.

In recent years, some of the most emerging applications in multimedia data processing are wireless/mobile multimedia systems and streaming content over the Internet. Both applications require flexible image data compression for storage or transmission proposals. Wavelet-based image compression schemes, such as the Embedded Zerotree Wavelet, obtain excellent results for these proposals and have been object of intensive research. In this work we propose an EZW-based compression method for colour images, based on the omission and restoration of wavelet subbands; our method achieves high compression rates and low computation times, combining, therefore, the advantages from both DCT and wavelet based compression algorithms.

The 3D measurement methods are very important in many fields. In particular, the methods based on 3D reconstruction from camera images are actively studied because these are passive detection methods and does not require expensive equipments. However, in order to achieve high accuracy, camera parameter calibration is very important. In this paper, we have proposed a new calibration method based on maximum likelihood estimation and confirmed the effectiveness by experiments.

In this paper we use a Multi-layer perceptron neural network with learning algorithm retropropagation errors, for application in fingerprint recognition. The objective is to measure the efficiency of the neural network by varying the test data. We observe the behavior of the network in the special case of a partial print. Once the overall structure of the network was designed, tested and properly trained, we proceeded with the testing process, varying the characteristic points and their particular characteristics. Overall, the results demonstrate a stronger recognition when all the characteristic points for the individual prints are available. The recognition rate begins to decrease as the number of characteristic points is reduced to 12, but increases when the number of points is 10, 8 or 5. We obtained a good percentage of hits to remove the features that depended on the center of the footprint and the footprint of the code, in this way to reach the desired goal.

This article describes a next-generation architecture for wireless communications, based on mobile phone carriers (GPRS) and broadband (WiFi), developed for the field of railways and enabling “train-to-earth” communications. This communication channel aims to complement traditional railway communication systems and its benefits make the deployment of new services, such as passenger oriented services, possible. Moreover, the result of this work is in itself a framework for the addition of new on-board applications that have the capacity to connect the trains with control points. As part of the architecture’s validation, currently underway, new digital services in the field of railways are being implanted.

Communication in multi-agent systems is an efficient way for solving cooperative tasks. Self-organizing communication help us understand the origin of the language and also it reduces designing time. In this paper, the effect of self-organizing communication in a foraging problem is studied. In our simulations a set of mobile robots which have to approach the food, construct a communication protocol to solve their task more efficiently. The effect of communication and individual perception on the performance of the team is analyzed.

We introduce a location information system for both indoors and outdoors which utilize WiFi location technology. The system is composed of a mobile terminal with a WiFi device and a communication server. We have developed seven location aware applications for the mobile terminal. Each of the application helps the user with current location information. We have performed a demonstration experiment in the subway of Nagoya City with 35 subjects and got a positive acceptance of the system.

Managing and sharing RDF triples in a distributed environment is one of the important issues for realizing semantic information retrieval in the Internet. Especially, the notion of RDF classes plays an important role in semantic information retrieval. In this paper, we propose a peer-to-peer information sharing method for RDF triples based on RDF class hierarchy. In the proposed method, a class ID is assigned to each RDF class so that it can represent inclusion relations of the class hierarchy. Then, each RDF triple is given a two-dimensional key that represents the classes of the subject and the object. By using Z-order of those two-dimensional keys, peers join an overlay network based on the Multi-Key Skip Graphs. Class-based retrieval of RDF triples is then realized by a range search on the overlay network. We show that the proposed method can process class-based RDF queries efficiently.

To enable efficient resource provisioning in HaaS (Hardware as a Service) cloud systems, virtual machine packing, which migrate virtual machines to minimize running real node, is essential. The virtual machine packing problem is a multi-objective optimization problem with several parameters and weights on parameters change dynamically subject to cloud provider preference. We propose to employ Genetic Algorithm (GA) method, that is one of the meta-heuristics. We implemented a prototype Virtual Machine packing optimization mechanism on Grivon, which is a virtual cluster management system we have been developing. The preliminary evaluation implied the GA method is promising for the problem.

The data-intensive applications are naturally executed on the Internet and generate a huge amount of data. The generated data would be stored distributed storages. In such environments, a user cannot easily find the target data by only filenames. The metadata is very useful to represent the characteristics and semantics of data. If users can specify the metadata, they will access the target data intuitively. We have been developing a distributed data management system called “MetaFa”. MetaFa can collect the metadata semi-automatically. In this paper, we discuss the implementation issues of MetaFa in order to collect metadata automatically.

In this paper, we propose the wireless LAN system using the airship. The wireless LAN system on the airship is tracked by the image processing automatically and provided the Internet connectivity to users on the ground in our proposed system. We describe the requirements of the wireless LAN system for the airship and design the architecture. To estimate the performance of our proposed system, we have performed the experiment with the airship flying in the real space. Experimental results indicate the validity of the wireless LAN system of image-based tracking antenna.

Ambient Intelligence has acquired great importance in recent years and requires the development of new innovative solutions. This paper presents a tele-monitoring system aimed at enhancing remote healthcare of dependent people at their homes. The system integrates distributed and heterogeneous Wireless Sensor Networks for optimizing the construction of ubiquitous scenarios. This approach provides the system with more flexibility to change its functionalities and components after the initial deployment than other analyzed proposals.

This work in progress describes an integral solution to implement a module-based platform for the control of all the automated systems inside a house. This project is based on the development of a biometrical access control system made by PAS Group in the University of Deusto. The major aim of this work is to develop an accessible way to control a house by the TV. By using the remote control, the clients will manage their houses. Also, it provides global autonomy to the disabled people, especially the old aged people. This will be made by knowing the physical restrictions of the collective who will live in the house. A set of different sensors will control this operation. In order to control all the home automated systems and coordinate the whole platform, this project will implement the OSGI standard. This implementation must ensure the system scalability in terms of enclosing new hardware. Also, it will be developed a ZibBee networking system to allow the communication between all the sensors, the systems and the central processor.

A system of supervision to disabled or aged people and enclosure access control has been developed. It integrates three types of sensors (acoustic, video and smart cards), and at the same time it is capable of interact with other safety systems. These system characteristics are: being robust, modular and extensible, as well as functional and ease to use. Its principal functionalities are: supervision of disabled people at his/her own housing by means of sensors of image (based on movement detection) and acoustic ones (included in a SODAR system), and users’ authentication by means of smart cards use.

Home telecare systems aim to support effective communications and emergency calls for people living in dependency situation but hardly provide reasoning capabilities to understand what to do in a problematic situation. This paper details the design and implementation of a reasoning platform to foresee or react in a smart way at home situations demanding care support for citizens from informal or remote carers. The system manages intelligent agents, whose behavior is defined and validated by ontologies and rules, to react in the elderly fall episode. A development methodology was adapted to sustain knowledge acquisition process from experts and to create the ontology for reasoning logic at homecare scenario. Thus, the platform is easily customizable to acquire data from telecare sensor networks, make reasoning according to each user profile and trigger ad hoc actions to communicate the problematic situation, to whom corresponds, or to interact with home appliances and residential gateways.

Smart homes for the aging population have recently started attracting the attention of the research community. One of the problems of interest is this of monitoring the activities of daily living (ADLs) of the elderly aiming at their protection and well-being. In this work, we present our initial efforts to automatically recognize ADLs using multimodal input from audio-visual sensors. For this purpose, and as part of Integrated Project Netcarity, far-field microphones and cameras have been installed inside an apartment and used to collect a corpus of ADLs, acted by multiple subjects. The resulting data streams are processed to generate perception-based acoustic features, as well as human location coordinates that are employed as visual features. The extracted features are then presented to Gaussian mixture models for their classification into a set of predefined ADLs. Our experimental results show that both acoustic and visual features are useful in ADL classification, but performance of the latter deteriorates when subject tracking becomes inaccurate. Furthermore, joint audio-visual classification by simple concatenative feature fusion significantly outperforms both unimodal classifiers.

In this paper we present the design of a low cost system which automatically controls the air conditioning equipments of a home. The sensors and controllers distributed use Zigbee, a protocol for wireless personal area networks, to communicate themselves. The aim of our design is to facilitate the development of complex smart homes applications, emphasizing the modularity of the system. This kind of applications is a great advantage for the end user, especially in the case of people with disabilities, who couldn’t interact with home’s electronic equipment.

During aging, older adults presents loss in their functional capabilities. This may cause that older adults do not continue performing their activities of daily living independently at home. We propose Ambient Information Systems (AIS) as the appropriate pervasive devices that can enrich the elders’ activities and promote their autonomy during the execution of their tasks. For illustrating this, in this paper we present an AIS for supporting medicine administration. By designing AIS as the presented in this paper, we have identified design issues and characteristics to be incorporated in AIS for supporting elder’s autonomy in their homes.

Ambient Assisted Living

(AAL) includes assistance to carry out daily activities, health and activity monitoring, enhancing safety and security, getting access to, medical and emergency systems. But home environments are challengeable as they’re different to represent. There are no only elderly people in a home In this paper we present the design of the contextual information for an intelligent home using a platform that exploits the modular and distributed architecture to develop context-aware applications.

A new regulatory framework, in Spain, has settled down a positive turn in social policies. Consequently, centres that provide special care and attention to handicapped persons have thrived. The staff attending such persons need to be skilled. The nature of both, staff and users, together with the health data treated, advises the implementation of a management information system that guarantees, controls and assures every care and service provided. The system presented in this paper intends to improve the workflow for the centres aforementioned, and assures every treatment that the users will be given thanks to RFID technologies. The social viability and economic feasibility of such centres is proven, reinforcing them is therefore an investment. The proposed system was tested in the Association of handicapped persons named

El Saliente

that through its

Centro Especial de Empleo

manages several day centres for elderly users and users with severe handicaps.

In this paper, we describe a radiology web system with an image analysis tool associated in order to measure bone density in fracture zones. These measures can be used to track the evolution of fractures in hip, knee, spine and long bones. It is being incorporated into the application a module that enables data mining. This module will induce a model for predicting the length of medical casualties based on patients’ data.

We present the main methodological guidelines followed during the Virtual Center for Elderly People project. This is a pilot experience of a telehealthcare platform for the elderly, capable of providing personalized medical care through the use of ambient intelligence. The focus of the project was the personalization of the service and the real-time knowledge generation from a distributed architecture including a sensor layer and computational models. The applied methodology is based on design-for-all philosophy, which we consider a successful approach to optimize cost-effectiveness, when the personalization is a requirement.

The increasing proportion of aged people in the population of developed countries requires the provision of assistance services based on remote continuous monitoring. Wireless sensors allow regular and real-time information to be obtained concerning health parameters in a non-intrusive way. The identification of critical values for these parameters and the computing possibilities of the current mobile devices provide support for a fast intervention which can minimize risks linked to delays in medical assistance. The diversity of individual situations has guided us towards a solution based on the software product line paradigm, as multiple options can be easily incorporated to the final product implementation. This article presents the product line generic architecture and an example of application, using a wireless sensor connected to a central station by means of a smart phone, which is able to detect alarm situations

This paper describes technology innovations towards computer aided memory management via intelligent data processing, and helping elderly people to overcome their decline in terms of cognitive. The system which integrates the functionalities to be delivered by HERMES, the FP7 funded project in Europe, aims at assisting the user who suffers from memory decline due to aging with effective memory refreshment based on the correlation of textual, spoken, or visual data. In this project, the system is being developed from a strong interdisciplinary perspective, which brings together knowledge from gerontology to software and hardware implementation.

Ambient Assisted Living (AAL) services are designed to cover the specific needs of elderly people. In a Smart Home environment many people can coexist requiring a continuous adjustment of the smart home services to their needs. The present work proposes a mechanism for the definition of self-adaptive AAL services that can dynamically reconfigure the Smart Home infrastructure without human intervention to fulfill the user requirements. Adaptation requirements are expressed in a declarative manner and AAL services are configured according to them. An OSGi-based infrastructure has been used to validate that the proposal can be applied in practice for an AAL context.

This paper describes an approach to user configuration of activity awareness. This approach offers users increased flexibility by allowing the use of multiple methods of configuration to be used within a unified framework; these configuration methods can include context, policy rules and automatic techniques of configuring behaviour. In this paper we describe the benefits of this flexibility and discuss a model of configuration that can be used to support these features.

User interaction with the intelligent environment should not require the user to adapt to special conventions or rules. It should be the environment who should adapt to the natural way of users interaction, but the tight resource constraints of the embedded sensors do not allow complex video processing algorithms to be executed in real time.

In this paper we present a low-cost approach to camera-based gesture recognition for intelligent environments, minimizing the required communication between sensors and servers, and performing most of the image processing in low-cost battery-powered microcontrollers.

HERMES aims at alleviating the elderly decline in both declarative and prospective memory, based on a set of ambient daily support and cognitive training applications. Its applications comprise memory aids and cognitive training games, both of which are presented in the paper. The paper introduces also the motivation of the HERMES project and applications. In addition it provides an overview of technical aspects in the areas of the project implementation in the areas of pervasive computing, as well as surface computing towards the utmost natural interactivity of seniors with computing devices.

Society is changing. People are older and need more services. Ambient Assisted is one the most important problem and a great studied field. In this field, we empathize systems that help people to make their daily activities. In this paper, we propose a system to control user daily behaviours and provide services according with them. We introduce our system in a Tagged World, a smart place that collects user activities using a sensor network. Our System uses information from sensors to extract behaviour sequence patterns that are used to reason and to obtain a specific service.

This paper covers those aspects of modern interfaces which expand and enhance the way in which people interact with computers, like multi-touch table systems, presence-detection led displays and interactive virtualized real-life environments. It elaborates on how disabled or conditioned people take great advantage of natural interaction as interfaces adapt to their needs; interfaces which can be focused towards memory, cognitive or physical deficiencies. Applications size-up to serve specific users with customized tools and options, and are aware while taking into account the state and situation of the individual.

The proportion of elderly people across the world is predicted to increase significantly in the next 50 years. Tools to assist the elderly with remaining independent must be developed now to reduce the impact this will have on future generations. Technological solutions have the potential to alleviate some of the problems associated with old age, particularly those associated with the deterioration of memory. This paper proposes an algorithm for semantic-based text processing within the context of a cognitive care platform for older people, and an implementation of the algorithm used within the EU FP7 project HERMES is introduced. The algorithm facilitates computerised human-like memory management through semantic interpretation of everyday events and textual search terms, and the utilisation of human language lexical resources.

Quality on the welfare services in caring, and the trend to minimize the economical and social-political costs that come with such practice, due to the population aging, are paramount nowadays, i.e., health care reform has become the leading policy issue in all latitudes. Indeed, the major thrust of all this research is the perception that escalating costs make the current structure and financing of health care unsustainable. The issue of sustainability is, therefore, the main subject of this paper. As a result, and in order to accomplish this goal, we decided to look to the problem from an user perspective, i.e., the system not only will provide different services, but will be also able to trace the ones more frequently used and to learn about the context in which they happen. As a result, we will have a system that will act and learn according to the preferences and habits of its users, and, simultaneously, will adapt to the environment with the objective of reducing the cost of its practices.

Real-time monitoring of human movements can be easily envisaged as a useful tool for many purposes and future applications. This paper presents the implementation of a real-time classification system for some basic human movements using a conventional mobile phone equipped with an accelerometer. The aim of this study was to check the present capacity of conventional mobile phones to execute in real-time all the necessary pattern recognition algorithms to classify the corresponding human movements. No server processing data is involved in this approach, so the human monitoring is completely decentralized and only an additional software will be required to remotely report the human monitoring. The feasibility of this approach opens a new range of opportunities to develop new applications at a reasonable low-cost.

Guaranteeing ubiquity and appropriateness of security and monitoring services provision to the users constitutes a priority issue for the authorities. This paper presents an innovative Wireless Personal Area Network architecture that takes advantage of some of the features provided by Intelligent Environments -large number of devices, heterogeneous networks and mobility enhancement- in order to adapt and personalise ambient conditions to the user profile. This system is based on image processing and its main aim is to provide an AAL solution that is integrated with other control devices for the home to make everyday tasks easier for users.

We explore the use of mobile social network technology combined with modern mobile phone hardware as a platform for programming applications in the elder care area. An application that covers two use cases for outdoors monitoring and detecting disorientations of the elderly is introduced. The system leverages on standard mobile terminals (Android G1) equipped with GPS and compass devices and on LibreGeoSocial, a mobile social framework we are developing.

This paper describes a solution to the elderly collective needs of assistance for the easy access to any type of buildings. The main purpose of the project is to provide a system which contributes to make easier the life of those who have problems of autonomy in their life’s. Nowadays, this part of the population experiences some problems when they use the traditional keys to access their own homes. The project shows how to implement an access control system that will face up this question with new technologies and following ‘Design for all’ rules. The ICT’s (Information and Communications Technologies) used are biometry and Radiofrequency IDentification (RFID). This combination makes this work one of the most innovative in the state of the art. The user, which must carry with him a tag, will be identified by the RFID reader and then the system will verify the identity by biometric pattern. The biometric pattern used in this project is the palm’s veins.

Unexpected falls and/or heart attacks at home are one of the main accidents the elderly face nowadays. This work focuses on elderly people which yet are independent and live alone in their own house. In such cases, the mentioned accidents may prevent her to ask for help as it is possible that she may lose conscience or stay paralyzed at the floor. In this paper, it is shown how a rule based classifier, designed by using simple a priori knowledge, which incorporates elderly’s context information and simple adaptive mechanisms for this information, may be used to detect domestic accidents as quickly as possible.

In this paper we propose a technological infrastructure for providing home assisted living support to mildly to medium mobility impaired users (particularly elderly people). The infrastructure includes an audio sensor network, a localization module and a behavioral analysis module. We also present a possible application scenario for our system and an overview of the components that integrate our system.

Ambient Assisted Living provides support for people’s daily life and aims at improving their quality of life. A health monitoring service could be intended to address the needs of sick people. Patient monitoring by medical personnel is frequently supported by handheld devices receiving health-care information. Location of these mobile devices is necessary in order to communicate any information, and the search strategy to locate them becomes a challenging issue in comparison to networks with permanent connections. We address this problem from an application point of view considering a membership based communication system characterized by users following repetitive patrol patterns day after day. We identify these patterns to generate a history of network connections to decrease the time required to locate any device on the network. We propose a modification of the commonly used Random Walk strategy to setup new connections on ad-hoc networks taking advantage of the learned patrol patterns.

This article presents a system which allows interaction between a physically disabled person and his environment. It contributes to achieve an Ambient Assisted Living (AAL). This system is controlled by voluntary muscular movements, particularly the orbicular ones. These movements are translated into instructions which are sent by means of a wireless link to a mobile robot that executes them. This robot includes a video camera in order to show the environment of the route that the robot follows on its way to the user. It also includes a subsystem that contains light and sound signals. This system can aid people with reduced mobility, extending the time that older people and disabled people can live in their home environment, increasing their autonomy and their confidence.

As a result of the increment of population in countries of Europe, a lot of efforts from European Authorities are coming from. In our research we want to bring forward a suite of developments related to build a ubiquitous AAL (Ambient Assisted Living) environment. We consider that recent approaches are based on ad-hoc technologies so its application is in this context isolated just in one domain of application. Our approach addresses to a reliable services platform for heterogeneous devices integration. On this basis we want to consider as well, the underlying benefits that a Service-oriented platform is giving to us in our Ambient Assisted Living Applications.

Wireless and Actuators Sensor Networks (WSANs) are one of the key technologies for supporting many Ambient Assisted Living applications. WSAN applications development poses new challenges like dealing with diverse low-level programming abstractions and the heterogeneity of nodes with critical resource limitations. Middleware platforms can hide from final developers the complexity of managing different types of hardware and software variability by applying a Software Product Line approach. This paper proposes a

family

of middleware for WSANs that can be customized according to the constraints imposed by the particular device, network and applications.

This work in progress presents a technological solution which lets some disabled people live independently in their own home. The objective is to develop a technical assistance that supports to them in the control of schedules, supervision of routes to the work place, warnings and automatic alarms generation. In addition, it will allow that people to communicate by telephone with the person of the family or Association which supervises to them. The definition of the specifications has been carried out in multidisciplinary form with technicians, psychologists, instructors and relatives. The tests have been developed in houses guarded by the Foundation Syndrome of Down of the Basque Country and by the department of Independent Life of this Foundation.

This paper presents the portable UVa Neuromuscular Training System (UVa-NTS). It is a myoelectric real-time system for research and upper-limb training. A set of training tools is included: this paper focuses on the game Myo-Pong, a simple graphical table-tennis game included in the UVa-NTS. To measure the performance, a set of control parameters is explained. Thus, Myo-Pong demonstrates the capabilities of the UVa-NTS as a myoelectric real-time system for training and for playing by means of myoelectric control.

In this paper, we present a proposal for Patients’ Mobile Monitoring. This framework enables the definition and generation of profiles, modules and communication structures between each of the measuring devices and the mobile phone depending on the kind of condition and the measuring values of the patient. We use patterns to allow the generation of self-control modules and patient profiles. These patterns establish relations between each module. With patient’s measured data, patient profile and modules, the framework generates an application for the doctor and the patient in a mobile phone. These applications allow the monitoring, patient self-control and the communication between the patient and the doctor. Moreover, as an important study case, we present a mobile monitoring system which allows patients with diabetes to have a constant control of their glucose tendency as well as direct communication with their doctor.

Alzheimer’s disease, the most common form of cortical dementia, is a degenerative brain disease for which there is no known cure but only a symptomatic therapy. Experts estimate that 26.6 million people worldwide had Alzheimer in 2006, which would multiply by four by 2050. The scope of the present paper is to present ALADDIN,

α

technology pLatform for the Assisted living of Dementia elDerly INdividuals and their carers, which aims at supporting maintaining health and functional capability, providing the means for the self-care and the self-management of chronic conditions, providing added value to the individual, leveraging his/her quality of life, while at the same time supporting the moral and mental upgrade of both the patients and their carers, as well as enhancing the home-as-care environment through the provision of tools for frequent, unobtrusive monitoring, via the development of user-friendly ICT tools.

The hospital and home are ancillary places in the life of elderly people, people with high risk of health problems or patients who have had recently an operation. We propose an architecture for Ambient Assisted Living (AAL) and health environments that supports pre-hospital health emergencies, remote monitoring of patients with chronic conditions, and medical collaboration through sharing of health-related information resources. We use the CEN/ISO EN13606 standard for EHR (electronic health record) and transfer information between medical systems, so that we can be open to the hospital systems and use the knowledge from the state of the patient at home for a better diagnosis, include directly sensors measure in the patient EHR among other. This architecture also supports the most important technologies for Home Automation, so we can combine security, comfort and Ambient intelligence solutions with a medical system. Thereby, to improve the elderly people’s Quality of Life (QoL).

This work explores how integration Shape Memory Fabrics boosts the interaction of People with Disability (PWD) and environment, via special fabric. In this respect, the transmission of haptic information is performed using different materials, like Nitinol Alloys and Electro Active Polymers; and also, through the design of a vibration system in order to create a communication tool. A key concept in our work is that of the user’s perceptions, since they are aware of their needs better than anyone else.

In this paper we propose a new approach for mental e-health treatments named intelligent e-therapy (e-it) with capabilities for ambient intelligence and ubiquitous computing. The proposed e-it system supposes an evolution of cybertherapy and telepsychology tools used up to now. The e-it system is based in a knowledge base that includes all the knowledge related to the disorder and its treatment. We introduce the use of ontologies as the best option for the design of this knowledge base. We also present a fist e-it system for obesity treatment called etiobe.

This paper presents results obtained in an ongoing project that deals with home care assistance for the elderly and disabled. The problems faced in this project cover many disciplines and can be studied using different approaches. Nowadays, e-health constitutes a young and expanding area that uses new technological innovation methods for social assistance. Methods and techniques from the Artificial Intelligence field offer a broad range of ideas and points of view for solving the problem. In particular, information systems and intelligent agents are two perspectives that deserve further study. Information systems provide a formal knowledge representation that models important tasks such as concept description and decision making. On the other hand, intelligent agents provide a mechanism to implement a rational behavior. The combination of both perspectives offers a valid solution to our problems.

Elderly people usually have some disabilities that get worst with the years. Many times these disabilities difficult the tasks carried out in a normal independent life, as is the case with home tasks. In addition, about one fourth of the household accidents happen in the kitchen. Within the framework of a European project –Easyline plus-, we have developed a tool to evaluate the quality of life of elderly people based on kitchen activity, extracted from data provided by the appliances. Such a tool has found to be very useful for social carers in order to monitor elderly activity, and as an objective support for diagnosis of the evolution of the personal abilities and autonomy of the user.

This paper describes the design and implementation of an object tracking service for indoor environments. First, the wireless indoor location estimation technology is overviewed presenting advantages and disadvantages. Second, the methodology of the study is presented. To estimate the position we use clues inserted by location clue injectors of the system. In our architecture one of these injectors is a ZigBee sensor network. As location algorithm we have developed a method combining statistical techniques (particle filter) and proximity sensing (nearest neighbour) to get better efficiency. The results obtained show that a good precision and reliability can be achieved with a low-cost solution.

This paper is a project report from the ongoing European project Nexes. We provide an overview of lessons learned from Nexes and propose guidelines for using business process modelling in integrated care processes. We will in this paper focus on the methodology used to model and describe existing chronic care processes and pathways in the Norwegian St. Olav’s hospital. The authors have been involved in the modelling activities in Nexes as researchers. We have informally collected our observations and have tried to compare them to other research results. Our conclusion is that conventional modelling languages and methodologies such as BPMN are useful but need to be adjusted and adapted to healthcare environments before being optimally exploited.

Text classification is becoming an interesting research field due to increased availability of documents in digital form which is necessary to organize. The machine learning paradigm is usually applied to text classification, according to which a general inductive process automatically builds an text classifier from a set of pre-classified documents. In this paper we investigate the application of Bayesian networks to classify MedLine documents, where each document is identified by a set of MeSH ontology terms. Bayesian networks have been selected for their ability to describe conditional independencies between variables and provide clear methodologies for learning from observations.Our experimental evaluation of these ideas is based on the relevance judgments of the 2004 TREC workshop Genomics track.

Our understanding of biological systems is highly dependent on the study of the mechanisms that regulate genetic expression. In this paper we present a tool to evaluate scientific papers that potentially describe

Saccharomyces cerevisiae

gene regulations, following the identification of transcription factors in abstracts using text mining techniques. GREAT evaluates the probability of a given gene-transcription factor pair corresponding to a gene regulation based on data retrieved from public biological databases.

Biomedical ontologies provide a commonly accepted scheme for the characterization of biological concepts that enable knowledge sharing and integration. Updating and maintaining an ontology requires highly specialized experts and is very time-consuming given the amount of literature that has to be analyzed and the difficulty in reaching consensus.

This paper outlines a proposal for the development of automated processes for the enrichment of the Gene Ontology (GO) that will use text mining techniques and ontology alignment techniques to extract new terms and relations. We also identify the areas of GO whose level of detail is too low to answer the community’s needs at large. We have found that although GO’s content is well suited to the manual annotations, revealing the coordination between GO developers and GO annotators, there are 17 areas that would benefit from enrichment to support electronic annotation efforts.

With this work we hope to provide biomedical researchers with an extended version of GO that can be used ’as is’ or by GO developers as a starting point to enrich GO.

Biomedical literature is an important source of information for chemical compounds. However, different representations and nomenclatures for chemical entities exist, which makes the reference of chemical entities ambiguous. Many systems already exist for gene and protein entity recognition, however very few exist for chemical entities. The main reason for this is the lack of corpus to train named entity recognition systems and perform evaluation.

In this paper we present a chemical entity recognizer that uses a machine learning approach based on conditional random fields (CRF) and compare the performance with dictionary-based approaches using several terminological resources. For the training and evaluation, a gold standard of manually curated patent documents was used. While the dictionary-based systems perform well in partial identification of chemical entities, the machine learning approach performs better (10% increase in F-score in comparison to the best dictionary-based system) when identifying complete entities.

In this work the problem associated to the obtaining of protein patterns associated with certain cancer types starting from biomedical texts is presented. The research is based on the study of the application of text mining and retrieval techniques to biomedical texts and its adaptation to this problem.Our goal is to annotate a significant corpus of biomedical texts, select the more relevant ones and to train machine learning methods to automatically categorize them along certain dimensions that we have previously defined. The idea behind this project is to identify a group of proteins associated with different cancer types.

In Biomedical research, the ability to retrieve the adequate information from the ever growing literature is an extremely important asset. This work provides an enhanced and general purpose approach to the process of document retrieval that enables the filtering of PubMed query results. The system is based on semantic indexing providing, for each set of retrieved documents, a network that links documents and relevant terms obtained by the annotation of biological entities (e.g. genes or proteins). This network provides distinct user perspectives and allows navigation over documents with similar terms and is also used to assess document relevance. A network learning procedure, based on previous work from e-mail spam filtering, is proposed, receiving as input a training set of manually classified documents.

Microarray technology allows measuring the expression levels of thousands of genes providing huge quantities of data to be analyzed. This fact makes fundamental the use of computational methods as well as new intelligent algorithms. This paper presents a Case-based reasoning (CBR) system for automatic classification of microarray data. The CBR system incorporates novel algorithms for data classification and knowledge discovery. The system has been tested in a case study and the results obtained are presented.

Sample-based clustering is one of the most common methods for discovering disease subtypes as well as unknown taxonomies. By revealing hidden structures in microarray data, cluster analysis can potentially lead to more tailored therapies for patients as well as better diagnostic procedures. In this work, we present a novel method for automatically discovering clusters of samples which are coherent from a genetic point of view. Each possible cluster is characterized by a fuzzy pattern which maintains a fuzzy discretization of relevant gene expression values. Noise genes are identified and removed from the fuzzy pattern based on their probability of appearance. Possible clusters are randomly constructed and iteratively refined by following a probabilistic search and an optimization schema. Experimental results over publicly available microarray data show the effectiveness of the proposed method.

Logistic regression (LR) is a simple and efficient supervised learning algorithm for estimating the probability of an outcome variable. This algorithm is widely accepted and used in medicine for classification of diseases using DNA microarray data. Classical LR does not perform well for microarrays when applied directly, because the number of variables exceeds the number of samples. However, by reducing the number of genes and selecting specific variables (using filtering methods) great results can be obtained with this algorithm. On this contribution we propose a novel approach for fitting the (penalized) LR models based on EDAs. Breast Cancer dataset has been proposed to compare both accuracy and gene selection.

Oligonucleotide Microarrays have become powerful tools in genetics, as they serve as parallel scanning mechanisms to detect the presence of genes using test probes. The detection of each gene depends on the multichannel differential expression of perfectly matched segments against mismatched ones. This methodology posse some interesting problems under the point of view of Genomic Signal Processing, as test probes express themselves in rather different patterns, not showing proportional expression levels for most of the segment pairs, as it would be expected. The method proposed in this paper consists in isolating gene expressions showing unexpected behavior using independent component analysis.

We present a method of grouping documents with genetic algorithms, the groups are created from the tokens representing the document. The system select the tokens starting from the Goffman point, selecting an area of suitable transition making use for it of the Zipf law. The experiments are carried out with the collection Reuters 21578 and the genetic algorithm uses the new operators designed to find the affinity and similarity of the documents without having prior knowledge of other characteristics. The proposed method is an alternative to the methods of traditional clustering and the results show that genetic algorithm is robust, clustering the documents in the collection of documents efficiently.

We present a novel approach to cluster sets of protein sequences, based on Inductive Logic Programming (ILP). Preliminary results show that the method proposed produces understandable descriptions/explanations of the clusters. Furthermore, it can be used as a knowledge elicitation tool to explain clusters proposed by other clustering approaches, such as standard phylogenetic programs.

In this paper we present a methodology for evaluating the confidence in the prediction of a physicochemical or biological property. Identifying unreliable compounds’ predictions is crucial for the modern drug discovery process.This task is accomplished by the combination of the method of prediction with a self-organizing map. In this way, the method is able to segregate unconfident predictions as well as confident predictions. We applied the method to four different data sets, and we obtained significant differences in the average predictions of our segregation. This approach constitutes a novel way for evaluating confidence, since it not only looks for extrapolation situations but also it identifies interpolation problems.

We present a summary of a PhD thesis proposing efficient biclustering algorithms for time series gene expression data analysis, able to discover important aspects of gene regulation as anticorrelation and time-lagged relationships, and a scoring method based on statistical significance and similarity measures. The ability of the proposed algorithms to efficiently identify sets of genes with statistically significant and biologically meaningful expression patterns is shown to be instrumental in the discovery of relevant biological phenomena, leading to more convincing evidence of specific transcriptional regulatory mechanisms.

This paper presents a method for performing a robust association between the apneas and hypopneas recorded on a polysomnogram and the desaturations they cause. It is based on a structural algorithm that takes advantage of the fuzzy set theory to represent the medical knowledge on which it relies. The method aims to generate information that could serve as a starting point for gaining a deeper insight into the Sleep Apnea-Hypopnea Syndrome by means of data mining techniques. This has led to a sacrifice of sensitivity for specificity.

We have validated our proposal over 37 hours of polysomnographic recordings. 88% of the hypoventilations present in the recordings were associated with the desaturations they caused, presenting a rate of false associations of 0.86%.

A key topic in population genetics is modeling the effect of population size on patterns of genome evolution. Here, we analyze a simple genetic algorithm which population size varies in time. We call this algorithm the genetic survival algorithm (GSA) that differs from the classical simple genetic algorithm just in that the selection model implies absolute fitness evaluation i.e. the individual survival does not depend onto the mean population fitness. It is shown that GSA provides an adequate model to study the evolution of multi-drug resistance in HIV-1 virus under anti-retroviral therapy.

The problem of inferring ancestral genetic information in terms of a set of founders of a given population arises in various biological contexts. In optimization terms, this problem can be formulated as a combinatorial string problem. The main problem of existing techniques, both exact and heuristic, is that their time complexity scales exponentially, which makes them impractical for solving large-scale instances. We developed a new constructive heuristic and a tabu search method with the explicit aim of providing solutions in a reduced amount of computation time. Experimental results show that when the number of founders grows, our algorithms have advantages over the ones proposed in the literature.

In this paper we present the work in progress on LogCHEM, an ILP based tool for discriminative interactive mining of chemical fragments. In particular, we describe the integration with a molecule visualisation software that allows the chemist to graphically control the search for interesting patterns in chemical fragments. Furthermore, we show how structured information, such as rings, functional groups like carboxyl, amine, methyl, ester, etc are integrated and exploited in LogCHEM.

The analysis of genomes and proteomes of the various organisms allow us to observe its behaviour in the evolution of species. In this study, we focus our attention on a particular aspect of this analysis: the conservation of specific codon and amino acid repetitions in orthologous genes belonging to eukaryotic organisms that are representative of different stages of species evolution. Since it is known that these repeats in humans are the cause of various neurodegenerative diseases, among others, this study help explaining if there is conservation or repression of such repetitions in the specialization process, and if there is any relationship between these repetitions and diseases in advanced live beings.

The main goal of this work is to introduce a new SIS epidemic model based on a particular type of finite state machines called cellular automata on graphs. The state of each cell stands for the fraction of the susceptible and infected individuals of the cell at a particular time step and the evolution of these classes is given in terms of a local transition function.

Integration of different kinds of biological processes is an ultimate goal for whole-cell modelling. We briefly review modelling formalisms that have been used in Systems Biology and identify the criteria that must be addressed by an integrating framework capable of modelling, analysing and simulating different biological networks. Aware that no formalism can fit all purposes we realize Petri nets as a suitable model for Metabolic Engineering and take a deeper perspective on the role of this formalism as an integrating framework for regulatory and metabolic networks.

In Systems Biology, there is a growing need for simulation and optimization tools for the prediction of the phenotypical behavior of microorganisms. In this paper, an open-source software platform is proposed to provide support for research in Metabolic Engineering, by implementing tools that enable the simulation and optimization of dynamic metabolic models using ordinary differential equations. Its main functionalities are related with (i) phenotype simulation of both wild type and mutant strains under given environmental conditions and (ii) strain optimization tackling tasks such as gene knockout selection or the definition of the optimal level of enzyme expression, given appropriate objective functions. The central carbon metabolism of

E. coli

was used as a case study, to illustrate the main features of the software.

The major objective of metabolic engineering is the construction of industrially relevant microbial strains with desired properties. From an engineering perspective, dynamic mathematical modeling to quantitatively assess intracellular metabolism and predict the complex behavior of living cells is one of the most successful tools to achieve that goal. In this work, we present an expansion of the original

E. coli

dynamic model [1], which links the acetate metabolism and tricarboxylic acid cycle (TCA) with the phosphotransferase systems, the pentose-phosphate pathway and the glycolysis system based on mechanistic enzymatic rate equations. The kinetic information is collected from available database and literature, and is used as an initial guess for the global fitting. The results of the numeric simulations were in good agreement with the experimental results. Thus, the results are sufficiently good to prompt us to seek further experimental data for comparison with the simulations.

The recent developments in life sciences and technology have produced large amounts of data in an extremely fast and cost-efficient way which require the development of new algorithms, coupled with massively parallel computing. Besides, biologists are usually non-programmers, thus demanding intuitive computer applications that are easy to use by means of a friendly GUI. In addition, different algorithms, databases and other tools usually lie on incompatible file formats, applications, operating systems and hardware platforms. It is therefore of paramount importance to overcome such limitations, so that bioinformatics becomes much more widely used amongst biologists. The main goal of our research project is to unify many of these existing bioinformatics applications and resources (local and remote) in one easy-to-use environment, independent of the computing platform, being a concentrator resource tool with a friendly interface. To achieve this, we propose a tool based on a new, open, free and well-documented architecture called Biomniverso. Two main elements make up such a tool: its kernel (Omega), which supplies services specifically adapted to allow the addition of new bioinformatics functionalities by means of plugins (like Minerva, which makes easy to detect SNP amongst a set of genomic data to discover fraudulent olive oil), and the interface (Brigid), which allows even non-programmer laboratory scientists to chain different processes into workflows and customize them without code writing.

PathJam is a new comprehensive and freely accessible web-server application integrating scattered human pathway annotation from several public sources. The tool has been designed to be intuitive for wet-lab users providing statistical enrichment analysis of pathway annotation for a given gene list of interest. Results are displayed in several interactive and downloadable views (graphs, spread-sheets, etc.) facilitating the biological interpretation of the gene lists. Moreover, PathJam allows users to build their own gene sets files in order to use them in gene set enrichment-based analysis. Finally, a simplified version of PathJam has been also implemented as a widget and is currently available for CARGO users.

Due to the high pace that algorithms for scanning genome-wide datasets are produced, most of the software is quickly released. As a consequence, they usually lack in system portability and only provide a text-based user interface. BioCASE is an open-source tool developed to assist bioinformaticians in the task of producing software to perform any kind of computationally-expensive genome-wide scanning. BioCASE will produce efficient software with some other added features such as to be portable, to have a graphical user interface (GUI) and to be easily set up. A first version of BioCASE (http://bios.ugr.es/biocase) has been completed and it is being currently used in the design of a genome- wide tool to perform variable selection of genotype data sets based on multivariate association and transmission-disequilibrium tests.

The constant increase on the amount and heterogeneity of biological data sources impose a permanent pressure toward the development of computational solutions that can integrate and process all the data, and help give answers to arising biological questions. Besides the work already developed on information integration in computational biology, the novel Web2.0 and Web Semantic trends leverage the design of next-generation applications sustained by the web-as-a-platform principle. Grounded on this idea, this paper presents a service orchestration framework that, using existing web components, allows the user to create and execute their own research workflow relying simply on a normal web browser.

SNP-microarrays are able to measure simultaneously both copy number and genotype at several single nucleotide polymorphism positions. Combining the two data, it is possible to better identify genomic aberrations. For this purpose, we propose a Bayesian piecewise constant regression which infers the type of aberration occurred, taking into account all the possible influence in the microarray detection of the genotype, resulting from an altered copy number level. Namely, we model the distributions of the detected genotype given a specific genomic alteration and we estimate the hyper-parameters used on public reference datasets.

A workflow based on the Taverna Workbench® software and tentatively named WPSA was developed to perform a generic protein sequence analysis eliminating the need to cut and paste data throughout web applications. The program performs a homology search, a multiple sequence alignment and a phylogeny analysis using PHYLogeny Interface Package (PHYLIP). The workflow designed gives a fast and significant answer to the user about the input sequence entered taking between 5 to 10 minutes to run depending on the Internet connection and web services.

The genetic code is the translation table used by an organism to transform each nucleotide triplet (codon) into an amino acid. For any species, its genetic code can be predicted by comparing conserved codon sites in protein alignments of target and reference sequences. Here we describe the design and implementation of a general and improved dynamic pipeline to predict the genetic code for any genome/species, taking into account the physicochemical properties of the amino acids involved and the phylogenetic relationships of the organisms compared.

PhyloGrid is a new tool continuously in progress whose aim is two-fold: to reduce the technological divide that a partial scientific community with low computational skills has for accessing new powerful computational platforms such as Grid; and, to develop a user-friendly interface by means of a workflow integrated in a web-portal. MrBayes software has been selected for calculating the evolutionary history of the species which is being integrated into new tools that performs the execution of jobs. Finally, this new application has been tested on the Human Papillomavirus.

ZARAMIT is an information system capable of fully automated phylogeny reconstruction. Methods have been tailored to mitochondrial DNA sequences, with focus on subproblem partitioning. We have built exhaustive human mitochondrial phylogenies (approximately 5500 sequences) and detected problems in existing haplogroup hierarchies through data-driven classification.

Information on the project can be found on

zaramit.org

.

Using simulated data, we tested whether the AFLP technique can consistently be used to estimate accurate evolutionary trees. We generated DNA sequences with known phylogenies that were used to simulate the AFLP procedure.

In silico

AFLP fingerprints were used to estimate neighbor-joining AFLP-based trees. The comparison of the true tree to the AFLP trees obtained over a wide range of conditions indicates that AFLPs usually lead to the wrong phylogeny except when DNA sequences are closely related, typically below the taxonomic rank of species.

The suitability of a proteomic approach was explored to establish phylogenetic relationships among closely-related species. Decapoda penaeid shrimps were chosen as case study because these species have been widely studied and their phylogenetic relationships have been inferred by extensively validated methods, among which mitochondrial DNA-based PCR studies have provided relevant information Thus, MALDI-TOF mass spectrometry (MS) peptide mass fingerprinting (PMF) of arginine kinase was performed, this enzyme being selected from the sarcoplasmic proteome of such species due to the interspecific variability of their pI values. The presence or absence of selected peptides in the MS spectra was used as a molecular marker for phylogenetic analysis. Based on the cluster analysis of the MALDI-TOF PMF spectra obtained, a dendrogram was generated which could be validated with those obtained using DNA-based methods.

Pairwise global alignment scores are used to detect related sequences in genome and proteins. These scores are biased by the length and composition of the compared sequences, and the Z-value is used to estimate their statistical significance. The Z-value is computed using a Monte Carlo algorithm that requires a large number of pairwise alignments between random permutations of the sequences compared.

A different alignment score, the

normalized edit distance

, is independent of the sequence lengths, and it usually takes 2 or 3 standard alignment calculations. In this paper we study the relationship between the normalized edit distance and the Z-value, and propose a method to screen pairs of unrelated sequences, so that Z-value needs to be computed for a small percentage of sequence pairs. We apply this method to the comparison of proteins from

Saccharomyces cerevisiae

,

Escherichia coli

,

Methanococcus jannaschii

and

Haemophilus influenzae

, showing that Z-value has to be computed for less than 1% of all protein pairs.

The examination of straightforwardly definable discrete structures in nucleic acids and proteins turned out to be perhaps the most important development in our present knowledge and understanding the their form and function. These discrete structures are sequences of nucleotides and amino acid residues, respectively. Bioinformatics was born as the science of analyzing these sequences. The discretization of the biological information into easy-to-handle sequences of 4 or 20 symbols made possible the application of deep mathematical, combinatorial and statistical tools with enormous success. The tools, resulting from this process, changed our perception of genetics, molecular biology, and life itself.

Straightforward discrete structures can also be defined in the spatial descriptions of proteins and nucleic acids. The definition and examination of discrete objects, using the spatial structure of proteins instead of amino acid sequences would intercept spatial characteristics, that are more conservative evolutionary than the polypeptide sequences.

In the present work we analyze the Delaunay tessellations of more than 5700 protein structures from the Protein Data Bank. The Delaunay tessellations of the heavy atoms of these protein structures give certainly a more complex structure than the polymer sequences themselves, but these tessellations are still easily manageable mathematically and statistically, and they also well describe the topological simplicial complex of the protein.

Our main result is Table 1, describing the relation between van der Waals and covalent bonds in the edges of the Delaunay tessellation. Among other findings, we show that there is only a single one Delaunay tetrahedron in the analyzed 5757 PDB entries with more than 81 million tetrahedra, where all six edges of the tetrahedron correspond to atom-pairs in van der Waals distance, but none of them to atom-pairs in covalent distance.

We present a new model of synthetic genetic oscillator based on a typical motif with one positive and one negative feedback loop. The repressor is a ribozyme, rather than a protein, which acts post-transcriptionally binding and cleaving to target mRNA. The properties of the ribozyme simplify our genetic oscillator that involves only two genes, one mRNA and one activator protein, apart from the ribozyme. Moreover, the genetic oscillator generates limit cycle oscillations, essential condition for resist the effects of the stochastic fluctuations due to the inherent randomness of the chemical reactions. As example of operation, we have chosen parameter values that produce circadian period in both deterministic and stochastic simulations, and the effects of stochastic fluctuations are quantified by a period histogram and autocorrelation function. Such new biochemical network designs may yield both new behaviors and better understanding of cellular processes.

This paper proposes a novel algorithm for complete exact pattern-matching focusing the specificities of protein sequences (alphabet of 20 symbols) but, also highly efficient considering larger alphabets. The searching strategy uses large search windows allowing multiple alignments per iteration. A new filtering heuristic, named compatibility rule, contributed decisively to the efficiency improvement. The new algorithm’s performance is, on average, superior in comparison with its best-rated competitors.

An approach to the inverse protein folding problem is described which combines a simulated annealing algorithm with template matching using the Bellman criteria. Solutions to proposed target structures are found by iteratively constructing the most similar solution. The folding model is based upon the traditional 2D HP protein lattice with a modified Viterbi dynamic programming algorithm. Initial results of both the optimal folding problem and the inverse protein problem are presented.

The paper focuses on the application of fuzzy sets in fault detection. The objective is to detect faults to an industrial gas turbine, with emphasis on the imbalance occurred in the rotor of the gas turbine. Such a fault has a certain degree of uncertainty and an index based on fuzzy sets has been developed in order to provide a fault confidence degree (0 meaning no fault, 1 the fault has been detected by all the sensors). Experimentation has been carried out on three real industrial turbines and it has shown the reliability and effectiveness of the methodology.

In the process of smelting copper mineral a large amount of sulphuric dioxide (SO2) is produced. This compound would be highly pollutant if it was emitted to the atmosphere. By means of an acid plant it is possible to transform SO2 into sulphuric acid. However, there are certain situations in the process of smelting copper mineral, in which SO2 escape to the atmosphere. This would be avoidable if we exactly knew under which circumstances this problem is produced. In this paper we present a practical application of KDD process, with an evolutionary algorithm as Data Mining technique, to the chemical industry. With this technique we obtain rules that make possible the definition of procedures that should help to optimize the functioning of the sulphuric acid production system. By means of the obtained results we show the viability of using automatic classifiers to improve a productive process, with decrease of the environmental pollution.

Load forecasting plays an important role in modern utilities. However, further improvements can be expected by predicting the load at a consumer level. The latter approach has become available with the advent of low-cost monitoring and transmission systems. Still, due to the limited number of monitored clients, the way groups of consumers should be identified and whether their data is sufficient for high quality prediction models remains an open issue.

The work summarises the results of building prediction models for different consumer groups of a district heating system. The way self-organising maps, multilayer perceptrons and simple prediction strategies can be applied to identify groups of consumers and build their prediction models has been proposed. The hypothesis that a billing database enables group identification has been verified. Significant improvements in prediction accuracy have been observed.

A novel variant of a multi-objective particle swarm optimization algorithm is reported. The proposed multi-objective particle swarm optimization algorithm is based on the maximin technique previously proposed for a multi-objective genetic algorithm. The technique is applied to optimize two types of problems: firth to a set of benchmark functions and second to the design of PID controllers regarding the classical design objectives of set-point tracking and output disturbance rejection.

This paper presents an automated design procedure of radio- frequency integrated CMOS discrete tuning varactors (RFDTVs). This new method use the maximin and the particle swarm optimization (PSO) algorithms to promote well distributed non-dominated fronts in the parameters space when a single-objective function is optimized. The fitness function used in the search tool is proportional to the RFDTV quality factor. The outcome of the automated design method comprises a set of RFDTV circuits, all having the same maximum performance. Each solution, which corresponds to one RFDTV circuit, is defined by the number of cells and by the circuit components values. This approach allows the designer to choose among several possible circuits the one that is easier to implement in a given CMOS process. To validate the effectiveness of the synthesis procedure proposed in this paper (PSO-method) comparisons with a design method based on genetic algorithms (GA-method) are presented. A 0.18

μ

m CMOS radio-frequency binary-weighted differential switched capacitor array (RFDSCA) was designed and implemented (the RFDSCA is one of the possible topologies of the RFDTVs). The results show that both design methods are in very good agreement. However, the PSO technique outperforms the GA-method in the design procedure run time taken to accomplish the same performance results.

The previous regulations about acoustics in buildings ruled only isolation, forgetting other important subjects. The new code in Spain, specifically the DB-HR document, also include the regulations of excessive noise reverberating that produces and causes discomfort in many cases non-speech intelligibility, this circumstance is crucial in certain areas. The passive control restrictions in these limits of frequencies are well known. The insulating materials, soundwalls, acoustic filters, Helmholtz resonator, expansion chamber, the encapsulation of the noise source, mean large dimensions and/or weight to 500 Hz or less. The active noise control (ANC Active Noise Control) can be used to cancel the noise to high-frequency and the passive techniques can be used to low-frequency. This paper presents algorithms for active noise control and the typical industrial applications.

This work presents a new method for license plate detection using neural networks in gray scale images. The method proposes a multiple classification strategy based on a Multilayer Perceptron. It consists of many classifications of one image using several shifted window grids. If a pixel belongs or not to the licence plate is determined by the most frequent answer given by the different classifications. The result becomes more precise by means of morphological operations and heuristic rules related to shape and size of the license plate zone. The whole method detects the license plates precisely with a low error rate under non-controlled environments.

In this paper, a trajectory tracking control for a nonholonomic mobile robot by the integration of a neural kinematic controller (NKC) and neural dynamic controller (NDC) is investigated, where the wheel actuator (e.g., dc motor) dynamics is integrated with mobile robot dynamics and kinematics so that the actuator input voltages are the control inputs, as well as both the kinematic and dynamic models contains parametric and/or nonparametric uncertainties. The proposed neural controller (PNC) is constituted of the NKC and the NDC, and were designed by use of a modelling technique of Gaussian radial basis function neural networks (RBFNNs). The NKC is applied to compensate the uncertainties in the kinematic parameters of the mobile robot. The NDC, based on the sliding mode theory, is applied to compensate the mobile robot dynamics, and parametric and/or nonparametric uncertainties. Also, the PNC are not dependent of the mobile robot kinematics and dynamics neither require the off-line training process. Stability analysis with basis on Lyapunov theory and numerical simulation is provided to show the effectiveness of the PNC.

Drilling is the most important operation in aeronautic industry carried out previous to riveting. Its main problem lies with the burrs. Nowadays, there is a burr elimination task (manual task) subsequent to drilling and previous to riveting that increases manufacturing cost. It is necessary to develop a monitoring system to detect automatically and on-line when the generated burr is out of aeronautic limits, and then deburring. This system would reduce holes deburring to the holes which really are out of tolerance limits, focusing in trying to avoid false negatives. The article shows an improvement in burr generation prediction, using Data Mining techniques versus current mathematical model. It gives an overview of the process from data preparation and selection to data analysis (with machine learning algorithms) and evaluation of the models.

This paper presents a neural recognition system for manufacturing applications in difficult industrial environments. In such difficult environments, where objects to be recognized can be dirty and illumination conditions cannot be sufficiently controlled, the required accuracy and rigidity of the system are critical features. The purpose of the real-time system is to recognize air-conditioning objects for avoiding deficiency in the manufactured process and erroneous identifications due to a large variety of size and kinds of objects. The architecture of the proposed system is based on several backpropagation neural networks in order to make an automatic recognition. Experimental results of a large variety of air-conditioning objects are provided to show the performance of the neural system in a difficult environment.

In this paper we present a soft computing system developed to optimize the face milling operation under High Speed conditions in the manufacture of steel components like molds with deep cavities. This applied research presents a multidisciplinary study based on the application of neural projection models in conjunction with identification systems, in order to find the optimal operating conditions in this industrial issue. Sensors on a milling centre capture the data used in this industrial case study defined under the frame of a machine-tool that manufactures industrial tools. The presented model is based on a two-phase application. The first phase uses a neural projection model capable of determine if the data collected is informative enough. The second phase is focus on identifying a model for the face milling process based on low-order models such as Black Box ones. The whole system is capable of approximating the optimal form of the model. Finally, it is shown that the Box-Jenkins algorithm, which calculates the function of a linear system from its input and output samples, is the most appropriate model to control such industrial task for the case of steel tools.