Distributed Computing and Artificial Intelligence

According to the progress of office automation, it becomes important to classify new and old bills automatically. In this paper, we adopt a new type of sub-band adaptive digital filters to extract the feature for classification of new and fatigued bills. First, we use wavelet transform to resolve the measurement signal into various frequency bands. For the data in each band, we construct an adaptive digital filter to cancel the noise included in the frequency band. Then we summarize the output of the filter output in each frequency band. The experimental results show the effectiveness of the proposed method to remove the noise.

In this paper we analyze the characteristics of an experimental otolith identification system based on image contours described with Elliptical Fourier Descriptors (EFD). Otoliths are found in the inner ear of fishes. They are formed by calcium carbonate crystals and organic materials of proteic origin. Fish otolith shape analysis can be used for sex, age, population and species identification studies, and can provide necessary and relevant information for ecological studies. The system we propose has been tested for the identification of three different species,

Engraulis encrasicholus

,

Pomadasys incisus

belonging to the different families (Engroulidae and Haemolidae), and two populations of the species

Merluccius merluccius

(from CAT and GAL) from the family Merlucciidae. The identification of species from different families could be carried out quite easily with some simple class identifiers -i.e based on Support Vector Machine (SVM) with linear Kernel-; however, to identify these two populations that are characterized by a high similarity in their global form; a more accurate, and detailed shape representation of the otoliths are required, and at the same time the Otolith identifiers have to deal with a bigger number of descriptors. That is the principal reason that makes a challenging task both the design and the training of an otolith identification system, with a good performance on both cases.

Availability of efficient mechanisms for selective and personalized recovery of information is nowadays one of the main demands of Web users. In the last years some systems endowed with intelligent mechanisms for making personalized recommendations have been developed. However, these recommender systems present some important drawbacks that prevent from satisfying entirely their users. In this work, a methodology that combines an association rule mining method with the definition of a domain-specific ontology is proposed in order to overcome these problems in the context of a movies’ recommender system.

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 overwork and exhaustion. A highly advanced bill classification technique, which distinguishes whether a bill is a reusable bill specifying the level of fatigue, is greatly required in order to comb out these problematic bills. Therefore, a purpose of this paper is to suggest a classification method of fatigue bills based on K-means with bill image data. The effectiveness of this approach is verified by the bill discriminant experimentation.

Recently the number of interactions between a company and its customers has been increased and it has taken a lot of time and cost of help desk operators. Companies construct FAQ pages in their web site and try to provide better services for their customer, however it takes surplus costs to analyze stored inquiries and extract frequent questions and answers. In this paper the authors propose a classification method of inquiry e-mails for describing FAQ (Frequently Asked Questions). In this method, a dictionary used for classification of inquiries is generated and updated automatically by statistical information of characteristic words in clusters, and inquiries are classified correctly to a proper cluster. This method achieved 70 percent precision of inquiry classification in an experiment with practical data stored in the registration management system for a sports association.

Recommendation systems can take advantage of semantic reasoning-capabilities to overcome common limitations of current systems and improve the recommendations’ quality. In this paper, we present a personalized-recommendation system, a system that makes use of representations of items and user-profiles based on ontologies in order to provide semantic applications with personalized services. The recommender uses domain ontologies to enhance the personalization: on the one hand, user’s interests are modeled in a more effective and accurate way by applying a domain-based inference method; on the other hand, the matching algorithm used by our content-based filtering approach, which provides a measure of the affinity between an item and a user, is enhanced by applying a semantic similarity method. The experimental evaluation on the Netflix movie-dataset demonstrates that the additional knowledge obtained by the semantics-based methods of the recommender contributes to the improvement of recommendation’s quality in terms of accuracy.

In this paper we present a hierarchical learning method to segment natural colour images combining the perceptual information of three natures: colour, texture, and homogeneity. Human knowledge is incorporated to a hierarchical categorisation process, where each nature features are independently categorised. Final segmentation is achieved through a refinement process using the categorisation information from each segment. Experiments are performed using the Berkeley Segmentation Dataset achieving good results even when comparing them to other significant methods.

Currently, a variety of eBooks with some intelligent capabilities to store and read digital books have been developed. With the use of these devices is easier to interact with the content available on the Web. But in some way access to such content is limited due to data overload problems. Trying to resolve this problem have been developed some techniques for information retrieval, among which are the recommender systems. These systems attempt to measure the taste and interest of users for some content and provide information relating to your profile. Through the feedback process attempts to collect the information that a recommendation system needs to work; but often this process requires the direct intervention of users, so that sometimes it is tedious and uncomfortable for users. For what we believe is necessary for a recommender system should be able to capture and measure implicitly the interaction parameters of a user with content in an eBook. Considering this need, we present a series of parameters that can be measured implicitly and how they will measured in the feedback process so that a recommender system to be reliable in electronic books.

Semantic web services have been studied during last years as an extension of Service-Oriented Computing on the Web 2.0. A lot of effort has been made to address some problems such as service discovery, matchmaking or service composition. Nevertheless, there is not much work in the literature about how to integrate trust into the process of selecting service providers. In this paper, an abstract architecture with a trust-based recommender module is presented, and a case study is put forward explaining how to apply the architecture to implement an agent in a mobile device.

This paper describes an agent’s movement protocol. Additionally, a distributed architecture to implement such protocol is presented. The architecture allows the agents to move in accordance with their requirements. The protocol is based on division and fusion of the agents in their basic components called Logical Sensors. The movement of the agents is based on the quality of services (QoS) and quality of control (QoC) parameters that the system can provides. The protocol is used to know the impact that the movement of the agents may have on the system and obtain the equilibrium points where the impact is minimal.

Nowadays, the society demands more and better products and services, and as a result organizations need to count on business strategies in order to provide a way for their employees communicate and fulfil their clients and work mates’ expectations more efficiently. This article presents the Hippocrates of Cos Multi-Agent System 1.0 (HdeC-MAS 1.0) a multi-agent system (MAS) based on virtual organizations. The system can access information and services ubiquitously from either land or mobile devices connected to wired or wireless networks. HdeC-MAS 1.0 supports the communication process for the healthcare services of a hospital. This paper presents the results obtained from the implementation of the system and demonstrates the advantages produced by the use of this new technology.

Multi-agent systems (MAS) paradigm emerged as an innovative technology that seemed to be applicable to a large number of distributed problems. However, during these years, ubiquitous computing and ambient intelligence among other distributed paradigms have proposed problems that are currently coped with other technologies. MAS have remained in research environments without establishing themselves in the distributed computing field, despite the benefits it could provide to it. In this paper, the key factors that have produced this situation are pointed, and solutions in order to fix it are proposed. APoDDS is a platform which collects all these solutions and merge them. Finally, a comparison between the new approach and the well-known agent platform JADE is made in order to evaluate the proposal.

This paper presents a multiagent architecture that covers the new requirements for the new control systems such as the distribution and decentralisation of system elements, the definition of communications between these elements, the fast adaptation in the control and organizational changes. The agents in this architecture can cooperate and coordinate to achieve a global goal, encapsulate the hardware interfaces and make the control system easily adapt to different requirements through configuration. Finally, the proposed architecture is applied to a control system of a solar power plant, obtaining a preliminary system that achieve the goals of simplicity, scalability, flexibility and optimization of communications system.

In this paper we introduce our future lines of investigation on the field of Intelligent Agents, focusing on analysing why agents which were well defined according to the guidelines of Agent Oriented Software Engineering, do not show the expected behaviour when they are introduced on a real environment, causing the system to collapse.When addressing to agents malfunctions we will use a metaphor in which agents are considered as patients waiting to be diagnosed. Problems agents show would be equivalent to diseases patients suffer from, and the diagnostic process in agents will also be compared to the procedure followed by physicians when they need to heal patients. Therefore we introduce the concepts of SithGent and SithGent Diagnose, to refer to, sick agents and its diagnostic process.

Nowadays there is a clear trend towards using methods and tools that can develop multi-agent systems (MAS) capable of performing dynamic self-organization when they detect changes in the environment. Moreover, the ideas that model the interactions of a multi-agent system cannot be related only to the agents and their communication skills in gaining strength, but it is necessary to use the concepts of organizational engineering. This paper presents a comprehensive list of development issues for primary adaptive virtual organizations (AVO). These issues will be a starting point for defining a complete list of AVO development requirements. From these requirements it could be possible to define an abstract architecture specifically addressed to the design of open multi-agent systems and virtual organizations.

Recent trends in communication technologies led to a shift in the already traditional Alternative Dispute Resolution paradigm, giving birth to the Online Dispute Resolution one. In this new paradigm, technologies are used as a way to deliver better, faster and cheaper alternatives to litigation in court. However, the role of technology can be further enhanced with the integration of Artificial Intelligence techniques. In this paper we present UMCourt, a tool that merges concepts from the fields of Law and Artificial Intelligence. The system keeps the parties informed about the possible consequences of their litigation if their problems are to be settled in court. Moreover, it makes use of a Case-based Reasoning algorithm that searches for solutions for the litigation considering past known similar cases, as a way to enhance the negotiation process. When parties have access to all this information and are aware of the consequences of their choices, they can take better decisions that encompass all the important aspects of a litigation process.

The paper describes a contextual environment using the Self-Organizing Map, which can model a semantic agent (SOMAgent) that learns the correct meaning of a word used in context in order to deal with specific phenomena such as ambiguity, and to generate more precise alignments that can improve the first choice of the Statistical Machine Translation system giving linguistic knowledge.

Virtual organisations are created to satisfy requests for complex services, after the creation phase they operate usually until they fulfil their objectives and dissolve the organisation freeing its members from their resource commitment towards the organisation; this is a common virtual organisation life-cycle. In some environments, the services requests may vary over time, having high numbers of requests at some periods requiring more organisations to cover them, resulting on high number of virtual organisations formation processes. But besides the fulfilment, other dissolution causes can be considered. In this paper we present other causes that should be considered, and explain how they can affect on the overall performance regarding the formation costs and services requests assignment. In addition, we present a virtual organisation test platform (VOCODIT, Virtual Organisation and COalition DIssolution Test platform) for evaluate this approach.

Cloud Computing presents a new approach to allow the development of dynamic, distributed and highly scalable software. For this purpose, Cloud Computing offers services, software and computing infrastructure independently through the network. To achieve a system that supports these characteristics, Service-Oriented Architectures (SOA) and agent frameworks exist which provide tools for developing distributed and multi-agent systems that can be used for the establishment of Cloud Computing environments. This paper presents a CISM@ (Cloud computing Integrated into Service-oriented Multi-Agent) architecture set on top of the platforms and frameworks by adding new layers for integrating a SOA and Cloud Computing approach and facilitating the distribution and management of functionalities. CISM@ has been applied to the real case study consisting of the analysis of microarray data and has allowed the efficient management of the allocation of resources to the different system agents.

In this study, a new monitoring system for carbon dioxide exchange is presented. The mission of the intelligent environment presented in this work, is to globally monitor the interaction between the ocean’s surface and the atmosphere, facilitating the work of oceanographers. This paper proposes a hybrid intelligent system integrates case-based reasoning (CBR) and support vector regression (SVR) characterised for their efficiency for data processing and knowledge extraction. Results have demonstrated that the system accurately predicts the evolution of the carbon dioxide exchange.

Simulation of plant community growth has become a hotspot in virtual reality area. In order to improve the level of knowledge sharing and solve the problem that the simulation parameters are difficult to adjust, a novel approach is presented to simulate the plant community growth, which is knowledge modeling based on competition. After analysing the related botany and ecology knowledge, this paper presents two models: intra-species competition model and inter-species competition model. Ontology, as a formal description of the plant competition, is adopted to express these two models. In addition, these models are deployed to the knowledge query prototype system for plant growth and different tree species have been simulated to validate these models.

Toward a really useful navigation system, utilizing spherical panoramic photos with maps like Google Street View is efficient. Users expect the system to be available in all areas they go. Conventional shooting methods obtain the shot position from GPS sensor. However, indoor areas are out of GPS range. Furthermore, most urban public indoor areas are crowded with pedestrians. Even if we blur the pedestrians in a photo, the photos with blurring are not useful for scenic information. Thus, we propose a method which simultaneously subtracts pedestrians based on background subtraction method and generates location metadata by manually input from maps. Using these methods, we achieved an underground panoramic view system which displays no pedestrians.

In this paper we describe a smooth switching mechanism for enhancing the performance and low-waste traffic of distributed peer-to-peer video streaming. The mechanism was designed for when streaming topology set up a backup link and a predicted link to avoid being congested link each nodes. Furthermore this provides a

Dominant Keyword

procedure which enables to improve the performance of switching time for changing from a peer to another peer. Finally we shows an implementation design and discuss about an efficiency of this proposal.

The high cost of existing industrial control systems prevents their implementation at small or medium size industrial plants. This paper describes a new architecture, based on low-power wireless sensor networks for remote monitoring and control of industrial equipment in small size plants. Through the use of very low consumption wireless devices, proposed architecture provides a low cost distributed control system easily deployable in small facilities.

This paper presents how the size of Textual Entailment Corpus could be increased by using Translators to generate additional 〈

t

,

h

〉 pairs. Also, we show the theoretical upper bound of a Corpus expanded by translators. Then, we propose an algorithm to expand the corpus size using Translator engines starting from a RTE Corpus, and finally we show the benefits that it could produce on RTE systems.

The optimal mapping of tasks of a parallel program onto nodes of a parallel computing system has a remarkable impact on application performance. We propose a new criterion to solve the mapping problem in 2D and 3D meshes that uses the communication matrix of the application and a cost matrix that depends on the system topology.We test via simulation the performance of optimization-based mappings, and compare it with consecutive and random trivial mappings using the NAS Parallel Benchmarks. We also compare application runtimes on both topologies. The final objective is to determine the best partitioning schema for large-scale systems, assigning to each application a partition with the best possible shape.

This paper explores the new opportunities offered by the emerging technologies of the last generation of mobile phones. Thanks to features like GPS facilities installed in the mobile terminals new value added services can be developed to offer the user the more suitable multimedia content depending on parameters like user’s location and preferences. We describe the development of a contextualized and personalized multimedia content delivery platform using transactional communications for mobile terminals. Furthermore an actual test route has been made to proof the successful working of the platform.

This paper discusses how to match the inquiry e-mails to pre-defined FAQs(Frequently Asked Questions). Web-based interaction such as order and registration form on a Web page is usually provided with its FAQ page for helping a user, however, most users submit their inquiry e-mails without checking such a page. This causes a help desk operator to process lots of e-mails even if some contents correspond to FAQs. Automatic matching of inquiry e-mails to pre-described FAQs is proposed based on SVM(Support Vector Machine) and specific Jaccard coefficient. Some experimental results show its effectiveness. We also discuss future work to improve our method.

Every time it is more often to audit the communications in companies to verify their right operation and to check that there is no illegal activity. The main problem is that the tools of audit are inefficient when communications are encrypted.

There are hacking and cryptanalysis techniques that allow intercepting and auditing encrypted communications with a computational cost so high that it is not a viable application in real time.

Moreover, the recent use of Graphics Processing Unit (GPU) in high-performance servers is changing this trend.

This article presents obtained results from implementations of brute force attacks and rainbow table generation, sequentially, using threads, MPI and CUDA. As a result of this work, we designed a tool (myEchelon) that allows auditing encrypted communications based on the use of hash functions.

A transcoder incorporating video watermarking method has been developed. This paper reports the implementation and evaluation of the system. Visibility testing showed that degradation in the quality of the watermarked images was almost imperceptible. Robustness testing showed that the embedded watermarks were robust against re-encoding with scaling. Process-time testing showed that the total processing time is increasing by 17%. Use of this system should help in deterring information leak in viewed and distributed video content.

Recently, the use of various chatterbots has been proposed to simulate conversation with human users. Several chatterbots can talk with users very well without a high-level contextual understanding.However, it may be difficult for chatterbots to reply to specific and interesting sentences because chatterbots lack intelligence. To solve this problem, we propose a novel chatterbot that can directly use Bulletin Board System (BBS) information in order to estimate user’s interests.

We wish to determine whether design engineers commonly use central projections (convergence of parallel lines to a vanishing point) when sketching new shapes, rather than draw physically parallel lines as parallel. This paper describes a pilot experiment carried out to determine the presence and importance of central projections. Results suggest that designers rarely use vanishing points when sketching engineering shapes. Hence, convergence can safely be ignored when designing and implementing basic artificial intelligence systems which detect perceptual cues in engineering design sketches. Since we wish to develop an automated method for discriminating between central and parallel pictorial projections, the paper also presents a numerical analysis of our results which could be used to calibrate such a method.

This paper presents an alternative detection method in fingerprinting technique for indoor localization and trajectory estimation based on efficient ray-tracing techniques over Wireless Local Area Networks (WLAN). Firstly, the use of radio frequency (RF) power levels and relative time delay is compared as detection method to estimate the localization of a set of mobile stations using the fingerprinting technique. The localization algorithm computes the Euclidean distance between the samples of signals received from each unknown position and each fingerprint stored in the database or radio-map obtained by using the FASPRI simulation tool. Secondly, an indoor trajectory has been simulated and tested by means of ray-tracing techniques. Experimental results shows that more precision can be obtained in the trajectory estimation by means of relative ray delay instead of RF power detection method, enabling the deploy of new applications for critical environments such as airports, where security and safety requirements are strongly involved.

Due to the increase of the technological area related to the radio localization, coverage study, both indoor and outdoor environments, the role of simulation tools rahter than traditional measurement techniques, are becoming more important. These tools are based on High Frequencies.

These are applied to analyze and optimize radio localization systems at indoor or outdoor scenarios, to characterize the channel propagation in these environments. A theoretical deterministic model based on the Uniform Theory of Diffraction is considered, in which the electrical field in an observation point is the sum of the contribution of all the rays that arrive to such point due the several propagation mechanisms like reflections, diffractions, transmissions and combinations of them.

Actual time series exhibit huge amounts of data which require an unaffordable computational load to be processed, leading to approximate representations to aid these processes. Segmentation processes deal with this issue dividing time series into a certain number of segments and approximating those segments with a basic function. Among the most extended segmentation approaches, piecewise linear representation is highlighted due to its simplicity. This work presents an approach based on the formalization of the segmentation process as a multiobjetive optimization problem and the resolution of that problem with an evolutionary algorithm.

In human-human interaction, a great deal of information is conveyed without explicit communication. This context information characterizes the situation of the different entities involved in the communication process (users, place, environment and computational objects). In this paper, we present an agent-based architecture that incorporates this valuable information to provide the most adapted service to the user. One of the main characteristics of our proposal is the incorporation of conversational agents handling different domains and adapted taking into account the different users requirements and preferences by means of a context manager. This way, we ensure a natural communication between the user and the system to provide a personalized service. The implementation of our proposed architecture to develop and evaluate a context-aware railway information system is also described.

The current interest and extension of social networking are rapidly introducing a large number of applications that originate new communication and interaction forms among their users. Social networks and virtual worlds, thus represent a perfect environment for interacting with applications that use multimodal information and are able to adapt to the specific characteristics and preferences of each user. As an example of this application, in this paper we present an example of the integration of conversational agents in social networks, describing the development of a conversational avatar that provides academic information in the virtual world of Second Life. For its implementation techniques from Speech Technologies and Natural Language Processing have been used to allow a more natural interaction with the system using voice.

This paper describes a context-aware mobile application which aims at adaptively motivating its users to assume active lifestyles. The application is built on a model which combines ‘motion patterns’ with ‘activity profiles’, in order to evaluate the user’s real level of activity and decide which actions to take to give advice or provide feedback. In particular, a ‘move-to-uncover’ wallpaper puzzle interface is employed as motivating interface; at the same time, context-aware notifications are triggered when low activity levels are detected. In order to accelerate the application’s design and development cycle, a mobile service oriented framework – CASanDRA Mobile - has been used and improved. CASanDRA Mobile provides standard features to facilitate context acquisition, fusion and reasoning in mobile devices, making easier access to sensors and context-aware applications cohabitation.

This paper outlines an architecture formulti-camera andmulti-modal sensor fusion.We define a high-level architecture in which image sensors like standard color, thermal, and time of flight cameras can be fused with high accuracy location systems based on UWB, Wifi, Bluetooth or RFID technologies. This architecture is specially well-suited for indoor environments, where such heterogeneous sensors usually coexists. The main advantage of such a system is that a combined nonredundant output is provided for all the detected targets. The fused output includes in its simplest form the location of each target, including additional features depending of the sensors involved in the target detection, e.g., location plus thermal information. This way, a surveillance or context-aware system obtains more accurate and complete information than only using one kind of technology.

This paper aims to present a new architecture to provide location services using multiple communication technologies such as Wifi, UWB, RFID and so on. Firstly, it will explain the advantages of multi sensor architecture against to use unique indoor location system and the reasons which led us to take this solution. Besides, this paper discusses the suitability of using ontologies for modeling message structure to locate in context-aware services platforms. This message will be described based on the concept of Asterix format used in aerospace multi-sensor communications.

In the present paper, we describes a semi-automated and decision support sensor surveillance architecture used to develop an intelligent sensor surveillance system. The proposed architecture is grouped in three agents layers: the sensors agents layer, sensor processing agents layer and finally, the support assistant agents layers. The sensor agents layer is formed by sensor managing agents and sensor data flow agents that they control the sensor devices and retransmit data streams to upper layer respectively. In sensor processing agents layer is an agents collection that process data flows produced by sensors, allowing elements tracking. The last layer is formed by special agents for helping and supporting the user monitoring and user choice. This architecture proposes a fully decentralized multi-agent system using FIPA Agent Communication Language.

Increasingly computer vision discipline needs annotated video databases to realize assessment tasks. Manually providing ground truth data to multimedia resources is a very expensive work in terms of effort, time and economic resources. Automatic and semi-automatic video annotation and labeling is the faster and more economic way to get ground truth for quite large video collections. In this paper, we describe a new automatic and supervised video annotation tool. Annotation tool is a modified version of ViPER-GT tool. ViPER-GT standard version allows manually editing and reviewing video metadata to generate assessment data. Automatic annotation capability is possible thanks to an incorporated tracking system which can deal the visual data association problem in real time. The research aim is offer a system which enables spends less time doing valid assessment models.

Ambient assisted living (AAL) services are usually designed to work on the assumption that real-time context information about the user and his environment is available. Systems handling acquisition and context inference need to use a versatile data model, expressive and scalable enough to handle complex context and heterogeneous data sources. In this paper, we describe an ontology to be used in a system providing AAL services. The ontology reuses previous ontologies and models the partners in the value chain and their service offering. With our proposal, we aim at having an effective AAL data model, easily adaptable to specific domain needs and services.

Face recognition is the most popular biometric used in applications at a distance, which range from high security scenarios such as border control to others such as video games. This is a very challenging task since there are many varying factors (illumination, pose, expression, etc.) This paper reports an experimental analysis of three acquisition scenarios for face recognition at a distance, namely: close, medium, and far distance between camera and query face, the three of them considering templates enrolled in controlled conditions. These three representative scenarios are studied using data from the NIST Multiple Biometric Grand Challenge, as the first step in order to understand the main variability factors that affect face recognition at a distance based on realistic yet workable and widely available data. The scenario analysis is conducted quantitatively in two ways. First, an analysis of the information content in segmented faces in the different scenarios. Second, an analysis of the performance across scenarios of three matchers, one commercial, and two other standard approaches using popular features (PCA and DCT) and matchers (SVM and GMM). The results show to what extent the acquisition setup impacts on the verification performance of face recognition at a distance.

Dynamic resource allocation according to user’s link quality is cirical in OFDMA system to improve network capacity. In this paper we consider joint subcarrier and power allocation of the downlink communication of multi-cell OFDMA system. The allocation problem is formulated with the goal of minimizing the transmitted power subject to individual rate constraint of the users. We propose a suboptimal distributed algorithm which consists of two stages. In the first stage each cell ignore the inter-cell interference and perform single-cell resource allocation. In the second stage the cells iteratively exchange the allocation result and update resource allocation until users’ rate requirements are met. The proposed algorithm is evaluated with computer simulations and compared with existing centralized algorithm. It is shown that the proposed algorithm obtain satisfactory tradeoff between quality of solution and complexity.

This paper presents a Genetic Programming algorithm which applies a clustering algorithm. The method evolves a population of trees for a fixed number of rounds or generations and applies a clustering algorithm to the population, in a way that in the selection process of trees their structure is taken into account. The proposed method, named DistClustGP, runs in a parallel environment, according to the model master-slave

,

so that it can evolve simultaneously different populations, and evolve together the best individuals from each cluster. DistClustGP favors the analysis of the parameters involved in the genetic process, decreases the number of generations necessary to obtain satisfactory results through evolution of different populations, due to its parallel nature, and allows the evolution of the best individuals taking into account their structure.

Currently, the best way to reduce the mortality of cancer is to detect it and treat it in the earliest stages. Automatic decision support systems are very helpful in this task but their performance is constrained by different factors and sometimes it is difficult to find a method with high sensitivity and specificity rates. One solution to this problem can be the collaboration between independent decision support systems. This article presents a proposal for a distributed and collaborative prostate cancer automatic diagnosis system based on artificial neural networks, which pretends to increase the accuracy of the decision support system combining the independent contributions of different artificial diagnosis entities.

The plethora of current positioning technologies, each one with very different features, together with the variety of environments wherein they are to be implanted, force system architects to thoroughly consider the choice for one of them in an isolated way, without combinining several options. Additionally, what makes a technology very appropriate in a certain constraints, may be the result of failing to fulfill others. Thus, trade-off solutions are usually to be made. In this paper, we provide a survey on different positioning techniques in relation to the satisfaction of certain non-functional requirements such as accuracy, responsiveness, complexity, scalability, etc, so that it can serve as guide to system designers in their ultimate decisions. The survey serves as an analysis and intends to highlight the need to undertake a new design capable of adapting this kind of distributed systems to specific characteristics of those technologies and environments; this objective could be achieved on the basis of a design considering non-functional such as requirements.

There are several different algorithms based on the ideas of collective behaviour of decentralized systems. Some of these algorithms try to imitate the distributed and self-organized systems that can be found in nature. Algorithms based on the mechanisms of distributed evidence gathering and processing of bee swarms are recent optimisation techniques. The distributed schema makes these algorithms suitable for a distributed implementation using the distributed computational infrastructures (DCIs) available. With these DCIs, large scale scientific problems can be optimized in a feasible time. However, the distributed paradigm of these infrastructures introduces several challenges in the design and development of any optimization technique. A distributed and asynchronous bees (DAB) algorithm running in a DCI is here presented with the aim to optimize any large scale problem.

Data mining has been widely applied to analyze data for decision makers. However, traditional data mining techniques are insufficient for analysis of multiple data sources. To mine multiple data sources, one possible way is reusing local data mining models discovered from each data source and searching for valid patterns that are useful at the global level. This paper presents a Knowledge Integration Model for integrating data mining models discovered from different data sources. This proposal is especially helpful for organizations which distributed data sources have been mined locally, and don’t share their original databases.

This research uses a Self-Organizing Map neural network model (SOM) as a short-term forecasting method. The objective is to obtain the demand curve of certain hours of the next day. In order to validate the model, an error index is assigned through the comparison of the results with the real known curves. This index is the Mean Absolute Percentage Error (MAPE), which measures the accuracy of fitted time series and forecasts. The pattern of input data and training parameters are being chosen in order to get the best results. The investigation is still in course and the authors are proving different patterns of input data to analyze the different results that they will be obtained with each one. Summing up, this research tries to establish a tool that helps the decision making process, forecasting the short-term global electric load demand curve.

The objective of the research is to prove the capability of Self-Organizing Map (SOM) to classify brake formula of a vehicle on a bank of roller tester from the MOT (Ministry of transport) and on flat ground. The neural network demonstrated good generation of the brake-slide relationship when presented with data not used in network training. This tool will easily find brake-slide equation of each experience and we will compare the brake on two different experimental tests. This article demonstrates that the MOT brake testing do not check the car brake in its usual way of driving. We will provide data and graphs to prove that tyre pressure is a determining factor when assessing the condition of brakes.

Clinical guidelines enable best medical evidence transfer to where best practice is needed. Although technology is considered the best way to reach this goal, the desired results have not been achieved yet.

In this work, we introduce a technological platform that allows the definition of guidelines including complementary information required by users. It is also ca-pable of generating platform-independent executable versions, thus improving the profitability of the undertaken effort. The systematisation of development using Model-Driven Development methods facilitates adaptation to changes and con-tinuous improvement of quality in both guidelines and infrastructure.

Developed guidelines, together with their browsable graphical representation, are made available to health professionals through our Web Portal e-GuidesMed.

After evaluating our guideline implementations on Rare and respiratory dis-eases, independent experts have emphasised their usefulness in daily practice and how valuable the technology is for supporting the development of new guidelines.

In this work we intend to advance towards a computational model to hold up a Group Decision Support System for VirtualECare, a system aimed at sustaining online healthcare services, where Extended Logic Programs (ELP) will be used for knowledge representation and reasoning. Under this scenario it is possible to evaluate the ELPs making in terms of the

Quality-of-Information (QoI)

that is assigned to them, along the several stages of the decision making process, which is given as a truth value in the interval 0...1, i.e., it is possible to provide a measure of the value of the

QoI

that supports the decision making process, an end in itself. It will be also considered the problem of

QoI

evaluation in a multicriteria decision setting, being the criteria to be fulfilled that of a Clinical Guideline (CG) for Chronic Obstructive Pulmonary Disease.

In this article a fuzzy system with capabilities for urological diagnosing is proposed. This system is specialized towards the diagnosis of urological dysfunctions with neurological etiology. For this reason the system specifies all the neural centres involved in both the urological phases, voiding and micturition. The fuzzy system allows to classify every dysfunction of all patients by means of their membership functions. The results of the experiments show that the fuzzy approach allows the diagnosis of urological dysfunctions from the relationship between neural centres and their associated neurological dysfunction.

In this paper we use a nonlinear processing technique based on mathematical morphology to develop a simple day trading system that automatically decides the timing to commute the marked strategy in terms of sort/long positions. In this short paper we show preliminary results.

This paper describes three different approaches based on complex heuristic searches to deal with a relevant telecommunication problem. Specifically, we have tackled a real-world version of the FAP –

Frequency Assignment Problem

by using three very relevant and efficient metaheuristics. Realistic versions of the FAP are NP-hard problems because the number of available frequencies to cover the entire network communications is always much reduced. On the other hand, it is well known that heuristic algorithms are very appropriate methods when tackling this sort of complex optimization problems. Therefore, we have chosen three different strategies to compare their results. These methods are: a very novel metaheuristic based on swarm intelligence (ABC –

Artificial Bee Colony

) which has not ever been used previously to tackle the FAP; a very efficient Genetic Algorithm (GA) which is a classical and effective algorithm tackling optimization problems; and one of the approaches that provides better results solving our problem: Scatter Search (SS). After a detailed experimental evaluation and comparison with other approaches, we can conclude that all methodologies studied here provide very competitive frequency plans when they work with real-world FAP, although the best results are provided by the SS and the GA strategies.

Multi-join query optimization is an important technique for designing and implementing database management system. It is a crucial factor that affects the capability of database. This paper proposes a Bees algorithm that simulates the foraging behavior of honey bee swarm to solve Multi-join query optimization problem. The performance of the Bees algorithm and Ant Colony Optimization algorithm are compared with respect to computational time and the simulation result indicates that Bees algorithm is more effective and efficient.

Continuous integration in software development is a practice recommended by the most important development methodologies. It promises many advantages such as early detection of bugs. An important element of continuous integration, although largely forgotten by the scientific literature, is the trigger, which initiates the process of building software from development sources. This paper discusses the possibility of improving this software component and opens the way for research that could be applied to other computer-related fields. To this end, we have implemented a prototype that shows for a case study, the results obtained when using existing triggers.

Advanced parallel Multi-Objective Evolutionary Algorithms (MOEA) have been used in order to solve a wide array of problems, including the planning of greenhouse crops. This paper shows the application of MOEA using the Island Parallel Model to solve a problem involving greenhouse crop planning in order to maximize profits and the production of biomass while reducing economic risks. The interest in maximizing biomass waste lies in the possibility of recycling it into heat and energy.

Antenna Positioning Problem (

app

) is an NP-Complete Optimisation Problem which arises in the telecommunication field. It consists in identifying the infrastructures required to establish a wireless network. Several objectives must be considered when tackling

app

and multi-objective evolutionary algorithms have been successfully applied to solve it. However, they required a deep analysis, and a correct parameterisation in order to obtain high quality solutions. In this work, a parallel hyperheuristic island-based model approach is presented. Several hyperheuristic scoring strategies are tested. Results show the advantages of the parallel hyperheuristic. On one hand, the testing of each sequential configuration can be avoided. On the other hand, it speeds up the attainment of high-quality solutions even when compared with the best sequential approaches.

The AntTree algorithm is a clustering method based on artificial ants which has been applied to document clustering, reaching good results. In this paper an improvement to the basic algorithm is proposed, based on the use of the information provided by the silhouette statistic. Computational results show that the improvement generates better results than the basic method.

The parameter values of a Multi-objective Evolutionary Algorithm greatly determine the behavior of the algorithm to find good solutions within a reasonable time for a particular problem. In general, static strategies consume lots of computational resources and time. In this work, a tool is used to develop a static strategy to solve the parameter setting problem, applied to the particular case of the Multi-objective 0/1 Knapsack Problem. GRID::Cluster makes feasible a dynamic on-the-fly setup of a secure and fault-tolerant virtual heterogeneous parallel machine without having administrator privileges. In the present work is used to speed-up the process of finding the best configuration, through optimal use of available resources. It allows the construction of a driver that launches, in a systematically way, different algorithm instances. Computational results show that, for a particular problem instance, the best behavior can be obtained with the same parameter values regardless of the applied algorithm. However, for different problem instances, the algorithms have to be tuned with other parameter values and this is a tedious process, since all experiments have to be repeated, for each new set of parameter values to be studied.

In this study, subjected to the trajectory generation for cooperative work, a genetic algorithm with cultural constructs is used to search for valid and optimal solutions in task space. We develop that algorithm by reflecting the behavior of social communities with a decision maker is used to evaluate cultural adaptation level by how well phenotypes, based on quaternion representation, are fitted in goal function. Algorithm uses cognition strategy to obtain smooth trajectory considering physical restrictive structure and actuator limits by using dynamic constrains in decision engine and eliminating unexpected derivation, also avoiding local minima problem.

Intrusion detection systems (IDS) are network security tools that process local audit data or monitor network traffic to search for specific patterns or certain deviations from expected behavior. We use a multiobjective evolutionary algorithm which is hybridized with an Artificial Immune System as a method of anomaly-based IDS because of the similarity between the intrusion detection system architecture and the biological immune systems. In this study, we tested the improvements we made to jREMISA, a multiobjective evolutionary algorithm inspired artificial immune system, on the DARPA 1999 dataset and compared our results with others in literature. The almost 100% true positive rate and 0% false positive rate of our approach, under the given parameter settings and experimental conditions, shows that the improvements are successful as an anomaly-based IDS when compared with related studies.

JXTA is a set of platform-independent, open source peer-to-peer protocols that has become popular for building services and applications based on peer-to-peer overlays. Existing work towards evaluating one of JXTA’s core protocols, the JXTA search algorithm, has mainly focused on testbed-based experiments. Although such evaluation configurations offer accurate results based on real deployments, scaling experiments to a large number of distributed hosts is difficult and often prohibitively expensive. Furthermore, repeating experiments using different configuration parameters might yield distorted results due to the uncontrolled nature of testbeds. Simulators offer an alternative to testbeds for evaluating large-scale applications in a controlled environment.This paper presents JXTA-Sim, a simulator for studying and evaluating the JXTA search algorithm. JXTA-Sim enables researchers to study the behavior of JXTA’s search algorithm using different configuration parameters and ultimately, to test JXTA-based peer-to-peer applications.

The development of an enhanced parallel algorithm for batch pattern training of a multilayer perceptron with the back propagation training algorithm and the research of its efficiency on general-purpose parallel computers are presented in this paper. An algorithmic description of the parallel version of the batch pattern training method is described. Several technical solutions which lead to enhancement of the parallelization efficiency of the algorithm are discussed. The efficiency of parallelization of the developed algorithm is investigated by progressively increasing the dimension of the parallelized problem on two general-purpose parallel computers. The results of the experimental researches show that (i) the enhanced version of the parallel algorithm is scalable and provides better parallelization efficiency than the old implementation; (ii) the parallelization efficiency of the algorithm is high enough for an efficient use of this algorithm on general-purpose parallel computers available within modern computational grids.

Particle Swarm Algorithm has demonstrated to be a powerful optimizer in multitude of optimization problems. The use of multipopulation technique with periodic interchange of individuals has proved to increase the convergence toward good solutions in many other EvolutionaryAlgorithms.However, the policy of interchange of individuals ought to be careful studied and selected, otherwise, pernicious effects could be introduced in the optimization process. The main focus of this study is on when, how and what individuals should be exchanged between populations in order to improve the convergence. In this paper, a deep study of diverse interchange policies for multipopulation applied to Particle Swarm Optimizer is presented.

The two-dimensional bin-packing problem (2D-BPP) with rotations is an important optimization problem which has a large number of practical applications. It consists of the non-overlapping placement of a set of rectangular pieces in the lowest number of bins of a homogenous size, with the edges of these pieces always parallel to the sides of bins, and with free 90 degrees rotation. A large number of methods have been proposed to solve this problem, including heuristic and meta-heuristic approaches. This paper presents a new memetic algorithm to solve the 2D-BPP that incorporates some operators specially designed for this problem. The performance of this memetic algorithm is compared with two other heuristics previously proposed by other authors in ten classes of frequently used benchmark problems. It is observed that, in some cases, the method here proposed is able to equal or even outperform to the results of the other two heuristics in most test problems.

Suppose Σ is the alphabet set and S is the set of strings with equal length over alphabet Σ. The closest string problem seeks for a string over Σ that minimizes the maximum hamming distance with other strings in

S

. The closest string problem is NP-complete. This problem has particular importance in computational biology and coding theory. In this paper we present an algorithm based on ant colony system. The proposed algorithm can solve closest string problem with reasonable time complexity. Experimental results have shown the correctness of algorithm. At the end, a comparison with one Meta heuristic algorithm is also given.

This paper proposes and studies the behavior of a new parallel cooperative model for trajectory based metaheuristics. Algorithms based on the exploration of the neighborhood of a single solution like simulated annealing (SA) have offered very accurate results for a large number of real-world problems. Although this kind of algorithms are quite efficient, more improvements are needed to address the large temporal complexity of industrial problems. One possible way to improve the performance is the utilization of parallel methods. The field of parallel models for trajectory methods has not been deeply studied. The new proposed parallel cooperative model allows both to reduce the global execution time and to improve the efficacy. We have evaluated this model in two very different techniques (SA and PALS) solving a real-world problem (the DNA Fragment Assembly).

This paper proposes the use of a parallelmultiobjective evolutionary technique to predict patterns, motifs, in real deoxyribonucleic acid (DNA) sequences. DNA analysis is a very important branch within bioinformatics, resulting in a large number of NP-hard optimization problems such as multiple alignment, motif finding, or protein folding. In this work we study the use of amultiobjective evolutionary algorithms team to solve the Motif Discovery Problem. According to this, we have designed a parallel heuristic that allows the collaborative work of four algorithms, two population-based algorithms: Differential Evolution with Pareto Tournaments and Nondominated Sorting Genetic Algorithm II, and two trajectory-based algorithms: Multiobjective Variable Neighborhood Search and Multiobjective Skewed Variable Neighborhood Search. In this way, we take advantage of the properties of different algorithms, getting to expand the search space covered in our problem. As we will see, the results obtained by our team significantly improve the results published in previous research.

Web cache performance has been reduced in Web 2.0 applications due to the increase of the update rate of the contents and of the personalization of the web pages. This problem must be minimized by the caching of content fragments instead of the complete web page. We propose a rule-based optimization algorithm to define the fragments design that experiment a best performance. This algorithm uses characterization parameters of the fragment contents to find the optimized solution.

Since the growth of the Internet,World Wide Web has become significant infrastructure in various fields such as business, commerce, education and so on. Accordingly, a user has gathered information by using the Internet. However due to increasing Web pages, it becomes difficult for a user to collect desirable information. Advanced Web search engines may provide solution to some extent, it is still up to a user to summarize or extract meaningful information from such retrieval results. Based on this viewpoints, this paper addresses a generation method of table-style data from heterogeneous Web pages that reflects a user’s intention. To achieve it, the method utilize a user’s instantiated example in a table in addition to column labels as the table. Based on a user’s instantiated example, meaningful information are extracted using pattern matching and N-gram method. We apply this method to 57 pages with 27 travel agencies whether the proposed method is effective or not. As the result, 88% was precision rate and 68% was recall rate.

Recently, since the growth of the Internet, WorldWide Web has become significant infrastructure in various fields such as business, commerce, education and so on. Accordingly, a user has gathered information by using the Internet. However due to the flood of Web pages, it becomes difficult for a user to collect desirable information. Advanced Web search engines may provide solution to some extent, it is still up to a user to summarize or extract meaningful information from such retrieval results. Based on this viewpoints, we addressed a generation method of table-style data from heterogeneous Webpages that reflects a user’s intention. However if original pages have less information, our system may not extract sufficient information. To improve this problem, we address a method that searches related page contents automatically.We apply this method to shopping sites and the experimental result shows it improves recall rate.

Ontologies are part of Semantic Web as models are part of Model-Driven Engineering, they can be seen as abstract, simplified views of the world. The possibility of transforming ontologies into software models, and vice versa, will bring both spaces together helping to achieve knowledge reuse. Both ontologies and models can assist in the domain analysis for the development of Domain-Specific Languages, so new transformations can be built to derivate DSLs from ontologies or models. This paper shows the current work in progress to build all these transformations and the concepts involved.

Recently, the use of various chatterbots has been proposed to simulate conversation with human users. Several chatterbots can talk with users very well without a high-level contextual understanding. However, it may be difficult for chatterbots to reply to specific and interesting sentences because chatterbots lack intelligence. To solve this problem, we propose a novel chatterbot that can directly use Web information. We carried out computational experiments by applying the proposed chatterbot to “2channel” (2ch) and “Twitter”.

With the vertiginous volume information growing, the amount of answers provided by traditional search engines and satisfying syntactically the user queries has enlarged directly. In order to reduce this problem the race to develop Semantic Search Engines (SSE) is increasingly popular. Currently, there are multiple proposals for Semantic Search Engines, and they are using a wide range of methods for matching the semantics behind user queries and the indexed collection of resources. In this work we survey the semantic search engines domain, and present a miscellaneous of perspectives about the different classification of approaches. We have created a comparative scheme and identified the prevalent research directions in SSE.

This paper reviews the different indexes and benchmarks used in the control performance assessment field of industrial processes. They are usually implemented to detect and diagnose malfunctions and disturbances in industrial controllers. This survey is just an overview of the methods and tools used in the control performance assessment/monitoring (CPA/CPM) technology which has been deeply studied over the last two decades.

Wavelength Division Multiplexing (WDM) in optical networks is the most favorable technology to exploit the huge bandwidth of this kind of networks. A problem occurs when it is necessary to establish a set of demands. This problem is called in the literature as Routing and Wavelength Assignment problem (RWA problem). In this paper we have used multiobjective evolutionary computing for solving the Static-RWA problem (demands are given in advance). We have implemented a population-based algorithm, Differential Evolution but incorporating the Pareto Tournament concept (DEPT). By using OpenMP, we have exploited the use of different multi-core systems (2, 4 and 8 cores), obtaining an average efficiency of 93.46% with our approach. To ensure that our heuristic obtains relevant results we have compared it with a parallel version of the standard algorithm NSGA-II. Furthermore we have compared the obtained results with other approaches and we can conclude that the DEPT algorithm has obtained better results.

Routing Problems arise in several areas of distribution management and logistics and their practical significance is widely known. These problems are usually difficult to solve. Therefore, heuristic methods are applied to try to solve them. This paper describes the application of artificial ant colonies to solve the General Routing Problem. For this, the problem is first transformed into a node-routing problem. The transformed problem is solved by applying an ant-based algorithm which has been widely applied to node-routing problems, obtaining good results.

We are developing an advanced IaaS (Infrastructure-as-a-Service) datacenter management system that dynamically minimizes running physical servers depending on resource utilization. The management system periodically monitors the loading of a datacenter, and dynamically repacks virtual machines (VMs) into optimal physical servers. Live migration of VMs and the standby mode of physical servers are automatically orchestrated by a genetic algorithm (GA) engine. A preliminary experiment showed that our first prototype system correctly worked for a proof-of-concept datacenter.

One of the useful patterns to create algorithms capable of solving complex problems is the foraging behavior of bees in finding food sources. In this article, a method has been presented for solving the complex problems in discrete spaces by simulation of this behavior of bees and also considering a memory for these bees. The proposed method has been successfully applied to solve the traveling salesman problem. The simulation results show the high ability of this algorithm in compare with the similar ones.

In this paper an iconic notation for describing the composition between the relations of a new qualitative representation model based on trajectories in two dimensions is presented. This qualitative representation model represents a new intuitive approach for describing the spatiotemporal features of two mobile entities through the relations between its trajectories. In order to describe the composition between relations in terms of a transitivity operation, an iconic notation derived from the Conceptual Neighborhood Graph is provided. Finally, in order to illustrate this iconic notation, some examples of composition between relations are presented.

We study the relation between a spectral deformation in speech processing and a geometrical deformation theory. We show that topological field theory yields the systematic treatment of these two methods. Some of the examples and the application to speech-spectra of classical mathematical ideas are discussed.

Electron tomography combines the acquisition of projection images through electronic microscope and techniques of tomographic reconstruction to allow structure determination of complex biological specimens. This kind of applications requires an extensive use of computational resources and considerable processing time because high resolution 3D reconstructions are demanded. The new tendency of high performance computing heads for hierarchical computational systems, where several shared memory nodes with multi-core CPUs are connected. In this work, we propose a hybrid parallel implementation for tomographic reconstruction of cellular specimens. Our results show that the balanced and adaptative algorithm allows an ideal speedup factor when large datasets are used.

Face recognition is an important biometric because of its potential applications in many fields, such as access control, surveillance, and human-computer interface. In this paper, we propose a rule-based face recognition system that fuses the output of two face recognition systems based on principal component analysis (PCA). One system uses the face image while the other use the Radon transform of the same face image. In addition, both systems use the Euclidean distance is the matching criteria. Both systems are trained using the same training images database, and fed with the same test input image at same time and the recognition result of each system is serving as input for the fusion decision stage. The proposed system is found to be better (97% recognition rate for recall and 93% for reject) than either system alone

The trace transform allows one to construct an unlimited number of image features that are invariant to a chosen group of image transformations. Object signature that is in the form of string of numbers is one kind of the transform features. In this paper, we demonstrate a wrapper method along with several ranking evaluation measurements to select useful features for the recognition of handwritten Jawi images. We compare the result of the recognition with those obtained by using methods where features are randomly selected or no feature selection at all. The proposed methods seem to be most promising.