Knowledge-Based and Intelligent Information and Engineering Systems

We have developed a table game named Innovation Game that supports users for thinking up ideas with social creativity. There are two types of players in the Innovation Game, innovators and consumers. While the innovators think up ideas and propose them to the consumers, the consumers criticize the ideas and make decisions whether they buy the ideas or not. In the Innovation Games, the innovators do not only propose their ideas to the consumers, but also improve the ideas using consumer’s comments that represent negative impression to the ideas. Therefore, it is considered that ideas with social creativity are related to the negative comments from the consumers. However, the relation between them has not been cleared. In this paper, we analyze discourse texts of communication generating social creativity. The analysis method focuses on the negative comments obtained from the consumers. We analyzed discourse texts of the Innovation Game using the method, and it was verified that the more negative comments the innovators accept, the more ideas with social creativity are obtained.

Value cognition system (VCS) is a human-centric system to enable value cognition, i.e., sensing, understanding, and taking advantage of latent values of entities. Here, human’s talents for value cognition are elevated and activated using tools such as sensors, software for social simulation and data visualization, etc., we will develop newly. The mechanism of this system will be characterized by the spiral process of four phases: (1)

sense

: experience scenes in the real world (2)

recollect

: recollect scenes relevant to a confronted situation, (3)

scenarization

: imagine scenarios to live/work with entities high-lighted via recollection and visualizing the data taken in (1), and (4)

co-elevation

: communicate the imagined scenarios and create a scenario to take advantage of values of entities. The scenario obtained shall be put into action, returning to step (1). Studies on chance discovery so far correspond to VCS applied for transient events.

The paper investigates a new temporal logic

${\mathcal LTL}^{\mathcal Z}_{{\mathcal D}{\mathcal U}}$

combining operations of the linear temporal logic LTL, the operation for discovery and operation for logical uncertainty. Our main aim is to construct a logical framework for modeling logical laws connecting temporal operations and operations of discovery and uncertainty. We consider questions of satisfiability and decidability for

${\mathcal LTL}^{\mathcal Z}_{{\mathcal D}{\mathcal U}}$

. Our principal result is found algorithm which recognizes theorems of

${\mathcal LTL}^{\mathcal Z}_{{\mathcal D}{\mathcal U}}$

(which implies that

${\mathcal LTL}^{\mathcal Z}_{{\mathcal D}{\mathcal U}}$

is decidable, and the satisfiability problem for

${\mathcal LTL}^{\mathcal Z}_{{\mathcal D}{\mathcal U}}$

is solvable).

Rule mining is considered as one of the usable mining method in order to obtain valuable knowledge from stored data on database systems. Although many rule mining algorithms have been developed, almost current rule mining algorithms only use primary difference of a criterion to select attribute-value pairs to obtain a rule set to a given dataset. In this paper, we introduce a rule generation method based on secondary differences of two criteria for avoiding the trade-off of coverage and accuracy. Then, we performed an evaluation of the proposed algorithm by using UCI common datasets. In this case study, we compared the predictive accuracies of rule sets learned by our algorithm with that of three representative algorithms. The result shows that our rule mining algorithm can obtain not only accurate rules but also rules with the other features.

Knowledge of living means information that people have obtained from their daily lives, and skills and wisdom that they have acquired through everyday experience or tradition. This paper aims to clarify three issues: (1) the uniqueness of living and scientific knowledge, (2) the significance of communication in these two kinds of knowledge, (3) the potentiality of a double helix structure for the two types knowledge in chance discovery, i.e. collaboration between lay subjects and specialists. These tasks were approached through theoretical and empirical research using concrete data obtained from a questionnaire and a case study. As a result, specialists and lay subjects were found to have outstanding knowledge in mutually different contexts even though they had limitations; solutions to problems were obtained through collaboration using mutual knowledge that was obtained on an equal footing.

This paper presents an integrated intelligent system being capable of automatically estimating and updating large-size input-output tables. The system in this paper consists of a series of components with the purposes of data retrieval, data integration, data analysis, and quality checking. This unique system is able to interpret and follow users’ XML-based query scripts, retrieve data from various sources and integrate them for the following data analysis components. The data analysis component is based on a unique modelling algorithm which constructs the matrix from the historical data and the spatial data simultaneously. This unique data analysis algorithm runs over the parallel computer to enable the system to estimate a large-size matrix. The result demonstrates the acceptable accuracy by comparing a part of the multipliers with the corresponding multipliers calculated by the matrix constructed by the surveys.

Context-aware applications allow service providers to adapt their services to actual user needs, by offering them personalized services depending on their current application context. Hence, service providers are usually interested in profiling users both to increase client satisfaction, and to broaden the set of offered services.

Since association rule extraction allows the identification of hidden correlations among data, its application in context-aware platforms is very attractive. However, traditional association rule extraction, driven by support and confidence constraints, may entail either (i) generating an unmanageable number of rules in case of low support thresholds, or (ii) discarding rare (infrequent) rules, even if their hidden knowledge might be relevant to the service provider. Novel approaches are needed to effectively manage different data granularities during the mining activity.

This paper presents the

CAS-Mine

framework to efficiently discover relevant relationships between user context data and currently asked services for both user and service profiling.

CAS-Mine

exploits a novel and efficient algorithm to extract generalized association rules. Support driven opportunistic aggregation is exploited to exclusively generalize infrequent rules. User-provided taxonomies on different attributes (e.g., a geographic hierarchy on spatial coordinates, a temporal hierarchy, a classification of provided services), drive the rule generalization process that prevents discarding relevant but infrequent knowledge.

Experiments performed on both real and synthetic datasets show the effectiveness and the efficiency of the proposed framework in mining different types of correlations between user habits and provided services.

In this paper we propose a system supporting the semi-automatic definition of inter-attribute mappings and transformation functions used as an ETL tool in a data warehouse project. The tool supports both schema level analysis, exploited for the mapping definitions amongst the data sources and the data warehouse, and instance level operations, exploited for defining transformation functions that integrate data coming from multiple sources in a common representation. Our proposal couples and extends the functionalities of two previously developed systems: the MOMIS integration system and the

RELEVANT

data analysis system.

This paper describes our continuing research on ontology-based knowledge source discovery on the Semantic Web. The research documented here is focused on discovering distributed knowledge sources from a user query using an Artificial Neural Network model. An experience using the Wordnet multilingual database for the translation of the terms extracted from the user query and for their codification is presented here. Preliminary results provide us with the conviction that combining ANN with WordNet has clearly made the system much more efficient.

This paper presents the initial steps followed in order to build an ontology about robot navigation (including specifically alternative navigation algorithms). Tackling the problem from general to specific, we start analyzing the desired behavior for generic mobile robots, in order to get common tasks and methods. Then, we fix our attention into the agricultural spraying robot developed in the University of Almeria by a multidisciplinary team. Because the field of robot navigation is consolidated, there are many algorithms (methods) to perform same activities (tasks). Our goal is to build an ontology including all this alternative methods, applying the dynamic selection of methods to make decisions in real-time depending on the environment conditions. Here we show the task-method diagrams with the parameterized description of some of alternative methods, using Fitorobot as testing case.

Self-managing databases intend to reduce the total cost of ownership for a DBS by automatically adapting the DBS configuration to evolving workloads and environments. However, existing techniques strictly focus on automating one particular administration task, and therefore cause problems like overreaction and interference. To prevent these problems, the self-management logic requires knowledge about the system-wide effects of reconfiguration actions. In this paper we therefore describe an approach for creating a DBS system model, which serves as a knowledge base for DBS self-management solutions. We analyse which information is required in the system model to support the prediction of the overall DBS behaviour under different configurations, workloads, and DBS states. As creating a complete quantitative description of existing DBMS in a system model is a difficult task, we propose a modelling approach which supports the evolutionary refinement of models. We also show how the system model can be used to predict whether or not business goal definitions like the response time will be met.

Spatial database systems built on top of distributed and heterogeneous spatial information sources such as conventional spatial databases underlying

Geographical Information Systems

(GIS), spatial data files and spatial information acquired or inferred from the Web, suffer from

data integration

and

topological consistency

problems. These issues make the globally-integrated spatial database

incomplete

, so that effectively and efficiently answering range queries over such databases represents a leading challenge for spatial database systems research. Inspired by these motivations, in this paper we propose

$\mathcal{I}$

-SQE (

Spatial Query Engine for

$\mathcal{I}$

ncomplete information

), an innovative query engine for answering range queries over incomplete spatial databases via meaningfully integrating

geometrical information

and

topological reasoning

.

$\mathcal{I}$

-SQE finally allows us to enhance the quality and the expressive power of retrieved answers by meaningfully taking advantages from the amenity of representing spatial database objects via both the geometrical and the topological level.

This paper examines an agent-mediated collaborative negotiation framework for e-commerce. This paper specifically focuses on travel industry. Individual customers and travel agencies will both be able to benefit from the usage of the system, since its negotiation strategies will not depend on price only, but several attributes, such as the number of rooms, the required facilities, and so on. The key issues in automating negotiation are the negotiation protocol, the negotiation object, and the negotiation strategy. Our paper addresses these issues by discussing the development of an agent-mediated e-commerce system using the FIPA compatible agent development framework, the JADE platform. Finally we provide our conclusions and discuss possible future work.

An agent-based system is characterised by an agent’s autonomous behaviour, which behaviour is the main difference between the concepts of agent and object. Agent autonomous behaviour is the ability of an agent to cooperate instead of integrate; therefore, the structure of agent-based systems consists of loosely coupled agents. In such an environment, the relationship between the agents is unlocked, so conventional, predefined integration software techniques are not an option because the agents need an open-architecture type of integration (cooperation) to achieve their tasks jointly. The aim of this research paper is to provide an evidence of how the ontology approach can play a role in modelling agent autonomous behaviour. The research explores the ontology software technologies used for semantic web applications, and designs a case study as an example of a set of services. In the implementation phase, the research uses the web ontology software development languages XML, RDFS, OWL, and Altova semanticWork to set up and develop the case study. The result is presented and plans for future work are discussed.

Security in wireless networks has become a major concern as the wireless networks are vulnerable to security threats than wired networks. The 802.11i wireless networks uses 4 way handshake protocol to distribute the key hierarchy in order to encrypt the data communication. In our previous research work [2], [3], we have investigated Quantum Key Distribution (QKD), for key distribution in 802.11 wireless networks. The whole communication flow of our proposed protocol can be split into several key processes. It can be seen that these processes can be implemented efficiently using Software Agents. In this paper we shall focus on the use of Software Agents in quantum cryptography based key distribution in WiFi wireless networks.

The problem considered in this paper is how to measure resemblance between images. One approach to the solution to this problem is to find parts of images that resemble each other with a tolerable level of error. This leads to a consideration of tolerance relations that define coverings of images and measurement of the degree of overlap between tolerances classes in pairs of images. This approach is based on a tolerance class form of near sets that model human perception in a physical continuum. This is a humanistic perception-based near set approach, where tolerances become part of the solution to the image correspondence problem. Near sets are a generalization of rough sets introduced by Zdzisław Pawlak during the early 1980s. The basic idea in devising near set-based measures of resemblance of images that emulate human perception is to allow overlapping classes in image coverings defined with respect to a tolerance

ε

. The contribution of this article is the introduction of two new tolerance class-based image resemblance measures and a comparison of the new measures with the original Henry-Peters image nearness measure.

Capillaroscopy is a branch of medicine which allows to diagnose various kinds of rheumatic diseases on the basis of observation of visual properties of nail-fold capillaries. Capillaries are tiny blood vessels of various shapes and sizes. Blood vessel tortuosity is one of medical signs. The paper presents a novel blood vessel tortuosity measure designed for capillary analysis. It represents the vessel as a graph and utilizes non-directional and directional traversal algorithms.

A new method of building local image features is proposed. The features are represented by various shapes (patterns) that can be approximated using Hough transforms. However, the transforms are applied locally (to the current content of a scanning window) so that the shape’s location is fixed at the current window’s position. Thus, the parameter-space dimensionality can be reduced by two (compared to globally computed Hough transforms) and the transforms can be effectively applied to more complex shapes. More importantly, shapes can be decomposed (two decomposition schemes are proposed) so that the overall complexity of the shapes used as features can be very high. The proposed feature-building scheme is scale-invariant (if scale is a dimension of the parameter space) subject only to diameters of scanning windows.

The growing importance of nail-fold capillaroscopy imaging as a diagnostic tool in medicine increases the need to automate this process. One of the most important markers in capillaroscopy is capillary thickness. On this basis capillaries may be divided into three separate categories:

healthy

,

capillaries with increased loops

and

megacapillaries

. In the paper we describe the problem of capillary thickness analysis automation. First, data is extracted from a segmented capillary image. Then feature vectors are constructed. They are given as an input for capillary classification method. We applied different classifiers in the experiments. The best achieved accuracy reaches 97%, which can be considered as very high and satisfying.

A huge number of different product types are managed through various processes in facilities with different approaches to scheduling. In this paper, we concentrate mainly on process-focused facilities. Sample groups of such facilities and processes were selected: its orders and times were investigated using both biological computation and genetic algorithms. First, biological computation was used to determine practical schedules. Second, genetic algorithms were used to identify which of the schedules determined by biological computation worked best. Here, we examine how combining these methods can be applied to solving process-focused scheduling problems.

In this paper, a new and systematic approach for the integration of fuzzy-based methods and biological computation, named as an evolutionary and biological method, is proposed for searching cliques in a fuzzy graph. When dealing with a number of nodes in a graph, the most intractable problem is often detecting the maximum clique, which is automatically obtained from finding a solution to the arranged cliques in descending order. The evolutionary and biological method is proposed to identify all the cliques and to arrange them in a fuzzy graph, and then to structure all the nodes in the graph, based on the searched cliques, in different hierarchical levels. This challenging approach, involving the integration of two techniques, provides a new and better method for solving clique problems.

Parallel processing functions using molecules have advantages to be exploited for classifying the given relational elements in a digraph. For instance, hierarchical structural modelling is used for classifying complicated objects into a hierarchical structure. In this paper, we consider the example of a digraph of hierarchical structural modelling that can be transformed to sequences of molecules, and propose a biologically intelligent method of encoding molecular sequences of different types, through the hierarchical classification of hierarchical structural modelling. Moreover, we show that this innovative biologically intelligent encoding method can be applied, not only to hierarchical structural modelling, but also to other relational problems composed of elements from digraphs.

Many of the applied methods and measurement tools of emotional usability engineering have been recommended for use designing products. A rough set method can also be a useful tool to be integrated with the basic concepts of emotional usability engineering. If such a method is applied, the groups of sensory words have to be investigated and their values reduced and classified to provide comprehensive information to product designers. However, a computational problem exists regarding the number of samples, groups of sensory words, and values required when resolving sense-based minimal decision rules. Considering this problem, we discuss the use of DNA computing, and propose a bio-inspired evolutionary method based on the rough set method, which should provide a new tool for emotional usability engineering.

A strategy for making layout decisions is an important element in developing operating systems in manufacturing factories or other industrial plants. In this paper, we look at fixed factory facilities and propose a method for designing different sorts of layouts related to factories running at high-volume and producing a low-variety of products. Where many tasks are called, each with a different task time, it can be difficult to arrange a fixed factory facility in the optimal way. Therefore, we propose a computational method using DNA molecules for designing production systems by determining all the feasible workstation groups in a fixed factory facility, and we show that this computation method can be generally applied to layout decisions.

In this paper, we present computational rule inferences to tackle the rate of aggregative risk in fuzzy circumstances. Based on the maximum membership grade principle, we apply the signed distance to defuzzify which is better than by the centroid. The proposed fuzzy assessment method is easier, closer to evaluator real thinking and more useful than the ones which have presented before.

A problem of worker assignment in cellular manufacturing (CM) environment is studied in this paper. The worker assignment problem is an NP-complete problem. In this paper, worker assignment method is modeled based on the principles of particle swarm optimization (PSO). PSO applies a collaborative population-based search, which models over the social behavior of fish schooling and bird flocking. PSO system combines local search method through self-experience with global search methods through neighboring experience, attempting to balance the exploration-exploitation trade-off which determines the efficiency and accuracy of an optimization. An effect of velocity controlled for the PSO’s is newly included in this paper. We applied the adaptation and implementation of the PSO search strategy to the worker assignment problem. Typical application examples are also presented: the results demonstrate that the velocity information is an important factor for searching best solution and our method is a viable approach for the worker assignment problem.

The objective of this study is to present an alternative approach to solve reasoning problems. DNA computing technique shows to solve evidential reasoning problems in this study. The reasoning is here executed on the basis of the concepts of plausibility and belief function. The evidential reasoning is a process dealing with problems that having both quantitative and qualitative criteria under various uncertainties including ignorance and randomness of information. The procedure to solve reasoning problem by means of DNA computing has been illustrated. An experiment shows the steadfastness of DNA computing to reach the solution of a reasoning problem.

Transportation plays a pivotal role in our society, especially in a good quality of life and economic prosperity. Intelligent transportation system (ITS) has been developed to manage the transport infrastructure and vehicles since the number of vehicles is rapidly growing and to avoid any accident. Various applications have provided to support ITS. One of them is a driver-assistant system. Considering of heavy vehicles such as bus, truck, trailer and etc., the driver assistant system is of importance in monitoring and recognizing objects in vehicle surrounding. For example, in operating a heavy vehicle, a driver has a limited view of the vehicle surrounding itself. It is difficult for the driver to ensure that the surrounding of vehicle is safe before operating the machine. Thus, in this paper, we employ a video tracking system through PSO and Parzen particle filter to break through several problems such as simultaneous motion and occlusion among objects. This method makes it easy to track a human movement from every frame and indirectly require less a processing time for tracking an object location in a video stream compared to conventional method. The detail outcome and result are discussed using experiments of the method in this paper.

Analyzing failure trends and establishing effective coping processes for complex problems in advance is essential in telecommunication services. We propose a method for semantically analyzing and classifying customer enquiries efficiently and precisely. Our method can also construct semantic content efficiently by extracting related terms through analysis and classification. This method is based on a dependency parsing and co-occurrence technique to enable classification of a large amount of unstructured data into patterns because customer enquiries are generally stored as unstructured textual data.

The mathematical framework for studying of a fuzzy approximate reasoning is presented. One of the defuzzification methods besides the center of gravity method which is the best well known defuzzification method is described. The continuity of the defuzzification methods and its application to a fuzzy feedback control are discussed.

This paper presents an agent-based simulator to investigating customer walking flows and purchasing behaviors in a super market. So far, such investigations have cost very much to examine in real situations. The simulator enables us to carry out “virtual experiments” through changing various parameters of retail businesses and store operations. For this purpose, first we observe an actual retail store and analyze sales data. Then we develop the simulation model: Agent-Based In-Store Simulator (ABISS). Intensive experiments have revealed that the flow of customers, which is related to the sales, depends on the design of a store and that the places of in-store advertisement and recommendation system vary their sales.

In text mining, importance indices of terms such as simple frequency, document frequency including the terms, and tf-idf of the terms, play a key role for finding valuable patterns in documents. As for the documents, they are often published daily, monthly, annually, and irregularly for each purpose. Although the purposes of each set of documents are not changed, roles of terms and the relationship among them in the documents change temporally. In order to detect such temporal changes, we decomposed the process into three sub-processes: automatic term extraction, importance index calculation, and temporal trend detection. On the basis of the consideration, we propose a method for detecting temporal trends of technical terms based on importance indices and clustering methods. By focusing on technical phrases, we carried out an experimentation to detect emergent and subsiding trends in a set of research document. The result shows that our method determined the temporal trends of technical phrases related to finding of patterns for innovations of research topics.

Collaborative filtering is used for the prediction of user preferences in recommender systems, such as for recommending movies, music, or articles. This method has a good effect on a company’s business. E-commerce companies such as Amazon and Netflix have successfully used recommender systems to increase sales and improve customer loyalty. However, these systems generally require ratings for the movies, music, etc. It is usually difficult or expensive to obtain such ratings data comparison with transaction data. Therefore, we need a high quality recommender system that uses only historical purchasing data without ratings. This paper discusses the effectiveness of a graph-partitioning method based recommender system. In numerical computational experiments, we applied our method to the purchasing data for CDs, and compared our results with those obtained by a traditional method. This showed that our method is more practical for business.

This research aims to present an analysis to optimally allocate advertising budgets based on single source data on consumers’ views of TV advertising. A model of consumer behavior and an optimality criterion for the advertising budget allocation are proposed together with a GA based optimization algorithm. Through the analysis, we discovered some knowledge to improve the effectiveness of advertising for several products.

We have studied query diffusion strategy to cover all the nodes over unstructured overlay network. Although our previous work [1] covers 80% of nodes over the power-law network, we step further to minimize the number of left-behind nodes. In order to propagate messages to overall network, we assume the best case to choose the shortest path between every node pair. This is aimed for studying the optimal message path as a whole, reducing the query diffusion cost which is equal to the sum of minimum shortest path length. We have studied the characteristics of message propagation behavior, and that our proposed strategy can be applied for contents delivery over unstructured overlay network.

This paper, we focus on recommendation functions to extract the high potential sales items from the trend leaders’ activities with the ID (Identification)-POS (Point-Of-Sales) data. Although the recommendation system is in common among the B2B or B2C businesses, the conventional recommendation engines provide the proper results; therefore, we need to improve the algorithms for the recommendation. We have defined the index of the trend leader with the criteria for the day and the sales number. Using with the results, we are able to make detailed decisions in the following three points: 1) to make appropriate recommendations to the other group member based on the transitions of the trend leaders’ preferences; 2) to evaluate the effect of the recommendation with the trend leaders’ preferences; and 3) to improve the retail management processes: prevention from the stock-out, sales promotion for early purchase effects and the increase of the numbers of sales.

Policy plans published by government institutions have been widely available to the public on the Internet. However, it has been noted that there often exist many unclear expressions in the sentences. In this paper, we have proposed a method of considering the level of ambiguity as relates to the amount of confidence the writer has in the implementation of the policy presented. In the policy plans analyzed in this study, many unclear expressions existed at the end of sentences. We also confirmed that such policies using unclear expressions lowered the rate of successful implementation as a whole. We also analyzed the appearance frequency of the expression “nado (and so forth)” which makes nouns and verbs within sentences unclear. As a result, we were able to confirm that the more expressions such as “and so forth” are used within one sentence, the lower the success rate of implementation of the policy becomes.

In system development projects, there is increasing number of cases requiring decision making to be conducted using official process. In this paper, the idea of Analytic Hierarchy Process and Linear Programming are introduced into the framework of the traditional decision making process. In addition, decision making process is continuously improved by eliminating cognitive bias in evaluator, by registering the differences between the initial decision making result and corrected result to Knowledge Database. The proposal was applied to a real case and the width of correction required has reduced after series of process improvements.

In the man-machine interfaces, it is important to use dialogue understanding technologies. One of the practical application fields is a question and answering (QA) systems. In order to reply appropriate answers for user’s questions, this paper presents a dialogue technique by transforming semantic expressions for both requests and answers. The measurements for the disrepute of the QA system are introduced for requests and answers, respectively. For the KAMOKUMA QA system generating answers which are reflecting user’s intension, the presented scheme is applied. For the AQ data with 7,518 requests, the real time simulation to estimate user’s sufficiency is computed.

The prospect of widespreading the Internet in business world is enabling new ways of solving our several business problems with many Q&A web sites. It is hard to say that good communication is carried out in these sites, because a lot of speakers may speak inconsistently in same site at same time. Therefore, we extracted the egocentrism from language expressions in words at some Japanese Q&A sites and tried to extract the gap of the mutual understanding in these words of business based on the egocentrism in this study. Specifically, we defined the weights for properties of presumed egocentrism with the method that presumes the egocentrism from language expressions in our previous study. We can presume the gap of the mutual understanding in the text data of business Q&A sites that have dialogue form by using the weights and calculating the strength score of the egocentrism in speech unit. We also evaluated this method with text data of business world in real Q&A sites and confirmed its effectiveness.

We have developed a new logic grammar for knowledge representation of electronic circuits. The grammar rules not only define the syntactic structure of electronic circuits, but also allow us to derive the meaning of a given circuit as relationships between its syntactic structure and basic circuit functions. In this paper, we show how voltage and current dependencies are coded in this new circuit grammar and how these dependencies are derived by parsing circuit structures.

Unfortunately there is no essentially faster algorithm than the brute-force algorithm for the nearest neighbor searching in high-dimensional space. The most promising way is to find an approximate nearest neighbor in high probability. This paper describes a novel algorithm that is practically faster than most of previous algorithms. Indeed, it runs in a sublinear order of the data size.

In preceding studies, error rate estimators have been compared under various conditions and in most cases the population distribution was assumed to be normal. Effects of non-normality of the population have therefore not been studied sufficiently. In this study, we focused on kurtosis as a measure of non-normality and examined the effects of kurtosis for error rate estimators, especially resampling-based estimators. Our simulation results in two-class discrimination using a linear discriminant function suggest that it is necessary to consider non-normality of the population in comparison of estimators.

Most organizations approach internal and external challenges with a varied degree of effectiveness. One of their biggest challenges is the ability to identify and respond appropriately to changes in their external environments. These changes affect not only their technological choices, but also their internal structures and cultures. In this context, we have seen an increasing demand for computational tools capable not only to support information storage but also to help in reasoning about the organizational environment. In particular, it is observed that the availability of a huge set of information in the Web offers a new opportunity to learn and reason about the organizational context. In this paper we present an empirical model to proceed the knowledge extraction from Web sources and support the reasoning process in the Competitive Intelligence domain.

In this paper, we propose a model that an agent selects actions based on variable neighborhoods. We formulate relationships among variable neighborhoods, the agent’s observations, and the agent’s behaviors in a framework of rough set theory and topological spaces. The main task is to explore a method by which we can select sizes of neighborhoods under given contexts. We also show simulation results of the proposed method.

Measuring and estimating are fundamental activities for the success of any project. In the software maintenance realm the lack of maturity, or even a low level of interest in adopting effective maintenance techniques and related metrics, have been pointed out as an important cause for the high costs involved. In this paper data mining techniques are applied to provide a sound estimation for the time required to accomplish a maintenance task. Based on real world data regarding maintenance requests, some regression models are built to predict the time required for each maintenance. Data on the team skill and the maintenance characteristics are mapped into values that predict better time estimations in comparison to the one predicted by the human expert. A particular finding from this research is that the time prediction provided by a human expert works as an inductive bias that improves the overall prediction accuracy.

The development of effective knowledge adaption techniques is one of the promising solutions to improve the performance of case-based reasoning (CBR) systems. Case-base maintenance becomes a powerful method to refine knowledge in CBR systems. This paper proposes an integrated knowledge adaption framework for CBR systems which contains a meta database component and a maintenance strategies component. The meta database component can help track changes of interested concepts and therefore enable a CBR system to signal a need for maintenance or to invoke adaption on its own. The maintenance strategies component can perform cross-container maintenance operations in a CBR system. This paper also illustrates how the proposed integrated knowledge adaption framework assists decision makers to build dynamic prediction and decision capabilities.

A paraconsistent annotated logic program called bf-EVALPSN has been developed for dealing with before-after relations between processes and applied to real-time process order control. In this paper, we propose a logical aticipatory system for before-after relation between processes based on reasoning of bf-EVALP vector annotations.

This paper is a sequel to [4]. We present an algebraic version of the monadic system P1* [8] by using the concept of Curry Algebra [5]. The algebraic structure obtained is called Monadic Curry Algebra P1*, which is a kind of ‘dual’ algebra studied in [4].

A new space-mapping method for object location estimation which is adaptive to camera setup changes in various applications is proposed. The location of an object appearing in an image is estimated by mapping image coordinates of object points to corresponding real-world coordinates using a mapping table, which is constructed in two stages, with the first for establishing a basic table using bilinear interpolation and the second for adapting it to changes of camera heights and orientations. Analytic equations for such adaptation are derived based on image formation and camera geometry properties. Good experimental results are shown to prove the feasibility of the proposed method.

A lateral vehicle localization method by omni-image analysis is proposed for car driving assistance. The method estimates analytically the position and orientation of a lateral vehicle by utilizing the geometric properties of a circular-shaped wheel image of the lateral car taken by a single omni-camera with a hyperboloidal-shaped mirror. Analytical solutions are made possible for fast computation by a special arrangement of affixing the omni-camera on the frontal car bumper at the height of the wheel. Experimental results showing good data estimation precision are included to prove the feasibility of the proposed method.

This paper proposes a novel patch-based approach for abnormal event detection from a mobile camera using concentric features. It is very different from traditional methods which require the cameras being static for well foreground object detection. Two stages are included in this system i.e., training and detection, for scene representation and exceptional change detection of important objects like paintings or antiques. Firstly, at the training stage, a novel scene representation scheme is proposed for large-scale surveillance using a set of corners and key frames. Then, at the detection stage, a novel patch matching scheme is proposed for efficient scene searching and comparison. The scheme reduces the time complexity of matching not only from search space but also feature dimension in similarity matching. Thus, desired scenes can be obtained extremely fast. After that, a spider-web structure is proposed for missing object detection even though there are large camera movements between any two adjacent frames. Experimental results prove that our proposed system is efficient, robust, and superior in missing object detection and abnormal event analysis.

Video inpainting is often used as a tool to assist user in removing objects or repairing damaged areas in a video. To deal with different kinds of video, several techniques such as object segmentation and temporal continuity maintenance are commonly adopted. In this paper, we extend the concept of exemplar-based inpainting to propose a new video inpainting algorithm which can realize object removal both in modern digital video and digitized aged films. Furthermore, a new patch searching strategy and a new patch adjustment mechanism are intorduced to maintain the temporal continuity of video and thus improve the results of video inpainting. Experiments demonstrate that the proposed algorithm can be effectively applied to different types of video.

This work develops a novel and robust hierarchical search tree matching algorithm, in which the Distance Transform based pedestrian silhouette template database is constructed for efficient pedestrian identification. The proposed algorithm was implemented and its performance assessed. The proposed method achieved an accuracy of 89% true positive, 92% true negative and low false positive 8% rates when matching 1069 pedestrian objects and 568 non-pedestrian objects. The contributions of this work are twofold. First, a novel pedestrian silhouette database is presented based on the Chamfer Distance Transform. Second, the proposed hierarchical search tree matching strategy utilizing Fuzzy C-means clustering method can be adopted for mapping and locating pedestrian objects with robustness and efficiency.

We propose a method that can detect pedestrians in a single image based on the combination of Adaboost learning with a local histogram features. Besides, instead of using the raw image for further processing, we introduce a layer enhanced by orientation filters which are superimposed to the original image. Experimental results obtained using the INRIA dataset show the superior performance of our method and thus demonstrate its robustness with the novel enhanced layer embedded pedestrian detector.

With rapid development of web 2.0 technology and e-business, bloggers play significant roles in the whole blogosphere as well as the external world. Specially, the most influential bloggers can bring great business values to modern enterprise in multiple ways, by increasing market profits and enlarging business impacts. The bloggers’ influences can be deployed only in a specific domain, e.g. computer companies only can utilize the influence bloggers’ expertise in computer knowledge, not their expertise in modern art or others. Despite that several influential bloggers mining systems are available, none of them consider the domain specific feature and their evaluations are based on generic influence, which is not applicable for real application requirements, such as business advertisement, personalized recommendation and so on. In this paper, we propose an effective model to mine the top-k influential bloggers according to their interest domains and network proximity. We investigate an effective algorithm to evaluate a blogger’s influence and develop a domain specific influential blogger mining system. The experiment results show that our system can effectively mine influential bloggers and is applicable to diverse applications.

Social network analysis offers many measures, which are successfully utilized to describe the social network profile. One of them is node position, useful to assess the importance of a given node within both the whole network and its smaller subgroups. However, to analyze large social networks a lot of effort and resources are necessary. In this paper, some algorithms that can be utilized in the process of node position evaluation are presented and their efficiency is tested. In particular, three distinct algorithms were developed and compared: PIN Edges, PIN Nodes, and PIN hybrid.

Reliable real data is indispensable for the examination, evaluation and the improvement of the organizational structure. This paper proposes a method to use patent documents for analyzing organizational structure of researchers. The method is more efficient and objective compared to personal interview. The structure of research groups is modeled as a “inventors graph”, which is a directed graph where each node represents an inventor and an edge represents co-inventor relationship. Empirical evaluation is conducted to cosmetic related companies and their patents that applied between 1998 and 2002 in Japan. It is shown that there is different characteristics in the inventors graph between Japanese companies and foreign companies. Moreover, time series analysis revealed that the inventors graphs of a Japanese company Kao changed in 2001 to foreign company type.

In this paper we propose an algorithm, which is an improvement of identification of web communities by [1], to extract research communities from bibliography data. Web graph is huge graph structure consisting nodes and edges, which represent web pages and hyperlinks. An web community is considered to be a set of web pages holding a common topic, in other words, it is a dense subgraph of web graph. Such subgraphs obtained by the max-flow algorithm [1] are called

max-flow communities

. We then improve this algorithm by introducing the strategy for selection of community nodes. The effectiveness of our improvement is shown by experiments on finding research communities from CiteSeer bibliography data.

Today, social networks systems have become more and more important. People have change their way to relate and communicate. Therefore, how to enhance contents and organization of a social network is a very important task. This way, we can help Virtual communities of practice (VCoP) to survive through time. VCoP are special kind of social network where the purpose is a key aspect. However, administrators are blind when trying to identify how to enhance the community. We propose a method which helps them by analyzing how purpose evolves through time. The approach has been experimentally tested in a real site with successful results.

In this paper, we present a method of discovering networks for modeling global propagation of influenza A (H3N2) viruses using a clustering algorithm. First, we find the clusters for every region by using an agglomerative hierarchical clustering with complete linkage. Next, we collect similar virus clusters over all regions. Finally, by comparing the occurrence year of the similar clusters, we construct a directed graph as a propagation network among these virus clusters.

The work presented in this paper addresses the application of new technologies to the task of intruder monitoring. It presents an innovative Machine Vision application to detect and track a person in a Closed Circuit Television System (CCTV) identifying suspicious activity. Neural Network techniques are applied to identify suspicious activities from the trajectory path, speed, direction and risk areas for a person in a scene, as well as human posture. Results correlate well with operator determining suspicious activity. The automated system presented assists an operator to increase reliability and to monitor large numbers of surveillance cameras.

Multiple, often conflicting objectives are specific to analog design. This paper presents a multiobjective optimization algorithm based on GA for design optimization of analog circuits. The fitness of each individual in the population is determined using a multiobjective ranking method. The algorithm found a set of feasible solutions on the Pareto front. Thus, the circuit designers can explore more possible solutions, choosing the final one according to further preferences/constraints. The proposed algorithm was shown to produce good solutions, in an efficient manner, for the design optimization of a CMOS amplifier, for two different sets of requirements.

Today there is a need for high performance chips that can provide very low power consumption, yet can operate over a number of application standards, such as operating a number of telecommunication standards depending on which country the device is in. This paper presents a new framework to enable the design of flexible systems by incorporating different range of reconfigurability in an embedded platform within an SOC design automatically. The SOC design automation involves identifying the best architectural features for the SOC platform, the configuration setting of reconfigurable cores, the type of interconnection schemes, their associated parameters such as data bandwidth, and placement of embedded cores in the communication infrastructures. For this optimization problem, a two-stage multi-objective optimization algorithm is presented. A multi-standard wireless telecommunication protocol is used to demonstrate our optimized designs in terms of area, power and performance.

This study elucidates issues related to using online vocabulary learning environments with collaborative filtering and functions for cognitive and social learning support in learner-centered learning, which requires learners to be self-regulated learners. The developed system provides learners with a vocabulary learning environment using online news as a test installation of functions. The system recommends news to each learner using a collaborative filtering algorithm. The system helps learners to use cognitive and social learning strategies such as underlining, along with a word-meaning display based on the learner’s vocabulary proficiency level. We investigated effects of the system on perceived usefulness and learning performance as a formative evaluation. Learners regarded this system as a useful tool for their language learning overall, but rated several functions low. Confirming the learning performance, the learner’s vocabulary proficiency level improved significantly.

The present paper reports the requirements, design, and learning effects of online collaboration support tools for project-based learning (PBL) applied to the development of embedded software. In this research, the authors created a new program that blends face-to-face classes and e-Learning classes. They also developed a computer-supported collaborative learning environment. In the present paper, the requirements for the collaboration support tools for the learning program are clarified through observation of a real PBL course. Based on this observation, an online repository tool and a unified search tool are proposed and implemented. The online repository tool was applied in a trial course blending face-to-face and online activities. Participants in the trial course completed a questionnaire survey. According to the survey responses, the blended learning program is feasible for PBL of embedded software design, and the online repository tool facilitates collaborative activities between learners and is effective for expanding each learnerfs design ability.

This study investigated learning styles of students who had or had not taken e-learning courses, developed a learning style questionnaire for e-learning courses, and examined the relationship between the learning style and the adaptability to e-learning courses. As the result, the student’s adaptability of e-learning courses can be suggested before his/her taking an e-learning course. It was found that using the multiple regression model obtained in the study, about 40% of the adaptability to e-learning courses can be explained by the learning style questionnaire developed in the study.

We propose a method to manage documents of engineering solution case based on TRIZ contradiction matrix theory. The document of engineering solution case involves know-how and techniques for solving mechanical issue. Usually the documents are written by engineers and practitioners, and managed by a company for sharing and inheriting among employees. However, an engineer who lacks literacy cannot find the previous case documents due to the inadequate keyword selections. To solve the query issue, we introduce TRIZ contradiction matrix theory for categorizing case documents. The engineers can retrieve adequate case documents by selecting improvement parameter and deterioration parameter on the matrix. Since the classification based on the matrix substantially categorize the case documents in terms of the problem solving methodology, it is effective and straightforward way of the specialized field and the key word. It is construction of the knowledge management support system that applies the idea of TRIZ. The problem solving that uses the reference information on this system is practiced and effectiveness is verified.

We have already developed an annotation sharing system for web-based learning materials. This system allows learners to write annotations such as markers and memorandums directly on materials and to share these annotations with lecturers and learners. If necessary, web-based materials can often be updated by authors; however, the annotations on the material are not in the proper position after the update. In the present study, we propose a method to follow the proper position of annotations in updated materials, and conduct experiments to evaluate the method. The following paper describes the proposed method and the experimental evaluation.

In collaborative learning, participants generate their own answers by exchanging their opinions through a discussion. Since the discussion in a collaborative learning includes knowledge for solving an exercise, the collaborative learning record is useful for other learners who tackle the same exercise. We propose a method for organizing solution knowledge in collaborative learning records as a solution knowledge graph. In this method, utterance collections of the same answering method are extracted and structured from a viewpoint of their effectiveness based on annotations attached by participants. In addition, the structure of the solution knowledge graph is refined by learning records of self-learners who use it as knowledge for solving exercises.

Problems involving agricultural know-how can be addressed with the use of IT. For example, IT can reduce the risk that know-how may be lost due to the increasing age of agricultural workers. Also valuable fruits which require sensitive environmental control can be monitored with IT. Data collection, collation and storage will enable us to convert tacit knowledge into formalized algorithms. We made a remote information sensing system using a sensor board made in cooperation with Renesas Solutions Corporation. Furthermore, we had installed our sensing system in a melon hothouse, in cooperation with the Prefectural Research Institute. We set up an interface enabling access to data and photographs using a browser. Using this interface, we expect that farmers will be able to transfer tacit knowledge into formalized information.

Clever management of the various types of documents used in intelligent activities and their efficient utilization are important. However, most available methods target only a single type of document (e-mails, Web pages, etc.). A more promising approach is topic-centered document management. Algorithms are described for extracting topics across various of types of documents. Moreover, a topic-centered document management system is described that is based on grouping by topics.

This paper describes a mechanism to annotate face images in impressive words which express their visual impressions. An annotation mechanism is developed by integrating latent semantic indexing, decision trees, and association rules. Moreover, visual and symbolic features of faces are integrated, which are corresponding to lengths and/or widths of face parts and impressive words, respectively. Relationships among these features are represented in a latent semantic space, their direct relationships in decision trees, and co-occurrence relationships among symbolic features in association rules, respectively. Efficiency of annotation results is improved by integrating these mechanisms, since their features are utilized effectively.

We developed an interactive learning environment for imitative figure sketching. Figure sketching is more difficult for novice than other sketching. People can easily find errors of figure sketch, since human is sensitive to human body figure. There are some important points to draw figure sketch. In this paper, we focus on length and angle between junctions. After learners draw human body figure by imitative sketching, the learning environment diagnoses the lengths and angles between junctions of drawn figure sketch. The environment shows scores of the learners’ figure sketch and some advice. We evaluated the environment with some learners.

We propose a distributed and autonomous algorithm for radio resource usage optimization in heterogeneous wireless network environment. We introduce optimization dynamics of the mutually connected neural network to optimize average throughput per the terminals and the load balancing among the radio access networks (RANs). The proposed method does not require a server to collect whole information of the network and compute the optimal state of RAN selections for each terminal. We construct a mutually connected neural network by calculating the connection weights and the thresholds of the neural network to autonomously minimize the objective function. By numerical simulations, we show that the proposed algorithm improves both the total and the fairness of the throughput per terminal. Moreover, we implement the proposed algorithm on an experimental wireless network distributively, and verify that the terminals optimize RAN selection autonomously.

There are many prior works of modeling travel behaviors. Most of them are investigated under the assumption that many kinds of data such as that of Person Trip (PT), which surveys travel behaviors, are available. Therefore, they do not consider an application to cities where the survey is not examined. In this paper, we propose a method for estimating travel behaviors using zone characteristics which is obtained from structural data of city. Focusing on dependent relationships between travel behaviors and city structure, we estimate the travel behaviors by means of the relationships. We first define trip and zone characteristics, and then introduce our method. With our method, we make use of Bayesian network constructed with PT data and the structural data. In addition, we show the effectiveness of our method through evaluation experiments.

Recently, the number of digital content objects is increasing rapidly with the progress of information technology. It has become important how we manage enormous digital content objects effectively and utilize them efficiently. Up to now, a lot of researches on digital content management have been reported. One of the important objectives of conventional management techniques is to search digital content objects that satisfy an information request of a user. This is based on the assumption that a user has one or more information requests. However, a user may have no information request explicitly when the user uses some kinds of devices for presenting digital content such as a digital photoframe. Such devices are expected to provide a presentation of digital content that rouses user’s interest. In this paper, we introduce an approach for composing photo slideshow that attracts user’s interest automatically.

To participate in an international conference, we must complete a series of tasks in accordance with the conference schedule. However, graduate students and researchers who have little or no experience of international conferences often have many difficulties in doing this. We propose a system to support their participation in international conferences. Our proposed system combines three functionalities, knowledge management, workflow management, and schedule management, and it takes advantages of several Web services. Our system associates knowledge, which is composed from the results and deliverables of performed tasks and related know-how, email messages sent by users, and their Web-search histories, with the tasks of conference workflows. In addition, when a workflow is created, the system adds important dates of the conference to the user’s Web calendar. We built a prototype system and confirmed that it works properly.

This paper analyzed the relationship between the evaluation of multi-agent systems and the agents’ environments. We define a movement difficulty for maps, using complexity indexes[1] and a vehicle movement simulator. In addition, we report on experiments carried out to confirm whether the movement difficulty can be used to estimate the results of the evaluation of agents. Finally, we investigate the similarity between the results of the simulations and the analysis.

Reprogramming sensor nodes is an effective way of improving wireless sensor networks. The latest reprogramming protocols use radio communication to distribute software data. Although several base stations are optimally placed to disseminate software rapidly in large-scale sensor networks, the performance of reprogramming protocols for multi-base station environments has not been discussed. This paper discusses our evaluation of the features of software dissemination by multi-base station sensor networks. Simulations revealed that the placement and number of base stations were the key parameters in software dissemination.

An on-demand bus system is now attracting attention as an alternative transport system for traditional fixed-route bus in Japan. In the on-demand bus system, buses transport customers door-to-door according to users’ demands, a user can freely specify the position of bus stop in its service area, and the desired time to get the buses. In this paper, we propose a model of the on-demand bus system using electrical vehicles and evaluate its feasibility by computer simulation. The characteristics of the electric vehicles are not considered in the past researches for on-demand bus problem. The improper charge timing decreases the acceptable rate of demands, and the lack of battery charge may occur while the vehicle is moving. In order to avoid such problems, we adopt the genetic algorithm to optimize transport plans. Simulation results showed that our transport model succeeded in the reduction of carbon-dioxide emissions by 80% and the running cost by 60% compared with traditional systems.

In this paper, we propose a framework for content collection and connection enabled by public large screens and mobile phones. Making people express their stories will encourage their active attitudes toward information management. With the proposed framework, we aim to overcome difficulties on managing flooding information. We applied the framework on two practice oriented systems and held workshops using them.

Multi-relational data mining (MRDM) is to enumerate frequently appeared patterns in data, the patterns which are appeared not only in a relational table but over a collection of tables. Although a database usually consists of many relational tables, most of data mining approaches treat patterns only on a table. An approach based on ILP (inductive logic programming) is a promising approach and it treats patterns on many tables. Pattern miners based on the ILP approach produce expressive patterns and are wide-applicative but computationally expensive because the miners search among large pattern space. We have been proposing a mining algorithm called MAPIX[3]. MAPIX has an advantage that it constructs patterns by combining atomic properties extracted from sampled examples. By restricting patterns into combinations of the atomic properties it gained efficiency compared with conventional algorithms including WARMR[1,2]. In order to scale MAPIX to treat large dataset on standard relational database systems, this paper studies implementation issues.

The process of creating photorealistic 3-dimensional computer graphic (3DCG) images is divided into two stages, i.e., modeling and rendering. Automatic rendering has gained popularity, and photorealistic rendering is generally used to render different types of images. However, professional artists still model characters manually. Moreover, not many progresses have been achieved with regard to 3-D shape data acquisition techniques that can be applied to facial modeling; this is an important problem hampering the progress of 3DCG. Generally, a laser and a highly accurate camera are used to acquire 3-D shape data. However, this technique is time-consuming and expensive. Further, the eyes may be damaged during measurements by this method. In order to solve these problems, we have proposed a simple method for 3-D shape data acquisition using a projector and a web camera. This method is economical, simple, and less time-consuming than conventional techniques. In this paper, we describe the setup of the projector and web camera, shape data acquisition process, image processing, and generation of a photorealistic image. We evaluate the error margin. We also verify the accuracy of this method by comparing the photograph of a face with its rendered image. After that, we pick up only labial and mouth part from obtained facial modeling data and expand it into animation.

In a Web study, using user-ID and password to specify learner is common. However, because of disguise problems etc., it is difficult to specify whether the learner on the terminal side is a person in question. Therefore, it is necessary to specify the learner by some other methods. In such methods, there are facial recognition, writer recognition, and other physical recognition methods such as fingerprint, iris, etc. The writer recognition is used for the handwriting analysis well as a concise procedure. In this research, the writer recognition was studied under the characters limited to hiragana. And the similarity differences by character types were examined. Moreover, the differences between similarity values obtained from own dictionary and others’ dictionaries were examined. As a result, introducing the stability of the character as one of the identification conditions was found to be necessary. And suitable characters for the writer recognition were obtained. From these results, a new writer recognition method was proposed.

We propose here an artificial vision model for the speed flexibility motion detection which uses analog electronic circuits and design the analog VLSI layout. In the previous model, the range of speed is quite narrow. However, we use the variable resistant parts inside the circuits. This model has speed flexibility property, and it is comprised of four layers. The model was shown to be capable of detecting a movement object. The number of elements in the model is reduced in its realization using the integrated devices. Therefore, the proposed model is robust with respect to fault tolerance. Moreover, the connection of this model is between adjacent elements, making hardware implementation easy.

This paper describes a new approach for self-calibration and color image rendering using radial basis function (RBF) neural network. Most empirical approaches make use of a calibration object. Here, we require no calibration object to both shape recovery and color image rendering. The neural network training data are obtained through the rotations of a target object. The approach can generate realistic virtual images without any calibration object which has the same reflectance properties as the target object. The proposed approach uses a neural network to obtain both surface orientation and albedo, and applies another neural network to generate virtual images for any viewpoint and any direction of light source. Experiments with real data are demonstrated.

Paintings have some sensibility information to human hearts. It is expected in paintings to process such sensibility information by computers effectively. For appreciation of paintings, grouping of paintings with similar sensitivity will be helpful to visitors as in painting gallery. In this paper, we developed a distance measure to group and classify similar paintings. Further, we applied the self organizing method (SOM) by two layered neural network to classify paintings. Then, the attributes of the sensibility of paintings are checked first. Next, color attributes of paintings are also checked. Paintings data with these attributes were computed by applying these techniques. Relatively well grouped results for the classification of paintings were obtained by the proposed method.

This paper presents a novel algorithm for localization in wireless sensor networks utilizing a fuzzy inference system at each sensor node. The algorithm using fuzzy distance measuring based on received signal strength information (RSS). The advantage of employing the RSS information is that no extra hardware is needed for localization. The simulation results and indoor experiments demonstrate that the proposed scheme employing fuzzy logic system can localize the mobile sensor nodes with certain accuracy.

WiMAX is an emerging technology based on the 802.16 standards to provide high speed and broadband wireless access for mobile stations. Handover is a key operation influencing the quality of communication services. In this paper, the scheme of choosing the suitable base station with the best service for a mobile station in a WiMAX network is studied. A new scheme based on fuzzy logic is proposed to employ the important traffic criteria, including bandwidth, dropping rate, blocking rate, and signal strength. Finally, the simulation is used to investigate the performance of proposed scheme. The simulation results show that the proposed schemes have better performance than conventional schemes, and can achieve the higher bandwidth utilization, the lower blocking rate for new calls, and the lower dropping rate for handover calls.

This paper applies bilevel optimization techniques and fuzzy set theory to model and support bidding strategy making in electricity markets. By analyzing the strategic bidding behavior of generating companies, we build up a fuzzy bilevel optimization model for day-ahead electricity market strategy making. In this model, each generating company chooses the bids to maximize the individual profit. A market operator solves an optimization problem based on the minimization purchase electricity fare to determine the output power for each unit and uniform marginal price. Then, a particle swarm optimization (PSO)-based algorithm is developed for solving problems defined by this model.

Fuzzy preference relations are frequently adopted by decision makers to express their preference tendency toward alternatives. Due to the lack of expertise of knowledge, decision makers may not be able to specify complete preference relation. To deal with incomplete fuzzy preference relations, Xu [26] proposed prioritization methods for incomplete fuzzy preference relations where he postulated a correspondence between priority vector and additive consistent incomplete fuzzy preference relation. In this paper, we are going to prove the correspondence does not always hold.

We present an approach for nature-inspired design of the driving style of an agent, remotely operating a scale model of a car with obstacle avoidance capabilities. The agent perceives the position of the car from an overhead video camera and conveys its actions to the car via standard radio control transmitter. In order to cope with the video feed latency we propose an anticipatory modeling in which the agent considers its current actions based on the anticipated intrinsic (rather than currently available, outdated) state of the car and its surrounding. Moreover, in a real-time the agent is able (i) to detect a static obstacle with a priori unknown coordinates using onboard video camera, (ii) to map the global position of the obstacle in a nature-inspired way by observing the dynamics of the change of visual angle (i.e., the motion parallax) of the obstacle in several consecutive video frames, and, (iii) in the vicinity of the latter, to employ a potential field-based obstacle avoidance maneuver. Presented work could be seen as a step towards the automated design of the control software of remotely operated vehicles capable to find a safe solution in changeable and uncertain environments.

In this paper, we propose a new decision support system for dealing stocks which utilizes the predictions (obtained by NNs) concerning the occurrence of the “Golden Cross (GC) and Dead Cross (DC)”, those (also obtained by NNs) concerning the rate of change of the future stock price several weeks ahead, and that (also obtained by NNs) concerning the relative position of the stock price versus “GC” and “DC”. Computer simulation results concerning the dealings of the TOPIX for the last 15 years confirm the effectiveness of our approach.

This paper considers an approach to fine grained parallel processing for soft computing that mainly deals with large-scale stochastic optimization problems. In the detailed steps of the computation, there are a lot of useless calculations that has no influence upon final results. Removing such a wasted process must be effective to reduce the computational cost. The key is asynchronization of data processing by using redundancy of variables and priority-based processing. A typical system architecture to support this approach is presented and discussed for its application.

In the last half century, systems such as those incorporated in electronic equipment and their constituent semiconductors coexisted favorably and supplemented each other while individually performing their own duties. In the 1990s, while low-price semiconductors with yet higher performance than those required by the systems became available, the market however started to see critical quality problems and the manpower and time required for system development increased. These problems are considered to be a result of the improper use of computer-aided design (CAD) that kept pouring an abundant supply of semiconductors into hardware logic circuits and program codes while failing to design what the systems should be like.

Presenting two systems which the authors are currently developing, this paper proposes a new method of designing effective systems while minimizing the costs necessary for system development, production and operation.

In AIDS treatments, it is an imperative problem to reduce the risk of the drug resistance. The previous study discussed which HIV-1 gene products are an ideal drug target not to develop drug resistance by applying some ideas of the graph theory, and suggested that the drug resistance would not develop if the drug target molecule functions as ”hub” in a chemical network where HIV-1 gene products interact directly or indirectly with intracellular agents in a HIV-1 host cell. The present study fortifies this suggestion in mathematical framework. The study develops the expression for a probability of drug resistance developing over the two different types: non-hub and hub of drug targets, and demonstrates that the hub drug target is more favorable for the drug resistance prevention than the non-hub one.

Seemingly conflicting phenomena of segregation and integration have been observed both in the immune system and the neural system, and possible mechanisms have been studied relating to learning and adaptation. Inspired by the

Stable Marriage Problem

whose solutions may exhibit both segregation and integration among agents, we propose a working model of matching automata which take order of preference as inputs and the resultant matching among agents as outputs. In this tentative model, we try to simulate the integration taking the switching experienced in the

Necker Cube

as an example, while the segregation is built into the restrictions of the model.

High-energy electron flux increases in the recovery phase after the space weather events such as a coronal mass ejection. High-energy electrons can penetrate circuits deeply and the penetration could lead to deep dielectric charging. The forecast of high-energy electron flux is vital in providing warning information for spacecraft operations. We investigate an adaptive predictor based on ADALINE neural network. The predictor can forecast the trend of the daily variations in high-energy electrons. The predictor was trained with the dataset of ten years from 1998 to 2008. We obtained the prediction efficiency approximately 0.6 each year except the first learning year 1998. Furthermore, the predictor can adapt to the changes for the satellite’s location. Our model succeeded in forecasting the high-energy electron flux 24 hours ahead.

Inspired by metabolic closure and its mathematical realization as a fixed point of

f(f) = f

where

f

is an operator, operand, and result, we pursue the possibility of reproduction closure of organisms. We seek an information aspect of reproduction closure, expecting an organizing principle of information (entropy) in living organisms. To remain reliable as a system with unreliable components, living organisms use reproduction involving the description (genotype). A reliability view of self-reproduction with a description will be compared to von Neumann’s complexity decrease principle in building automata. Asymmetry of complexity decrease would indicate that the self-reproduction in his model is not reversible, hence suggesting entropy generation (negative entropy leak). Although errors in copying description would lead to threats of cancer, allowance of a certain level of error would lead to possible adaptation recognizing the openness of biological systems.

Eudaemonics is the study of the nature, causes, and conditions of human well-being. According to the ethical theory of eudaemonia, reaping satisfaction and fulfillment from life is not only a desirable end, but a moral responsibility. However, in modern society, many individuals struggle to meet this responsibility. Computational mechanisms could better enable individuals to achieve eudaemonia by yielding practical real-world systems that embody algorithms that promote human flourishing. This article presents eudaemonic systems as the evolutionary goal of the present day recommender system.

Organizations’ integrate different systems and software applications in order to provide a complete set of services to their customers. However, different types of organisations are facing a common problem today, namely problems with security in their systems. The reason is that focus is on functionality rather than security. Besides that, security, if considered, comes too late in the system and software engineering processes; often during design or implementation phase. Moreover, majority of system engineers do not have knowledge in security. However, security experts are rarely involved in development process. Thus, systems are not developed with security in mind, which usually lead to problems and security breaches. We propose an approach of integration security throughout engineering process. To assure that necessary actions concerning security have been taken during development process, we propose semi-automated preventive controls.

User centricity required by ubiquitous computing is about making services and information be prepared and delivered in the perspective of users rather than system elements. Task-oriented computing supports user centricity by representing users’ goals in tasks. It bridges the gap between tasks and available services. This paper proposes a semantically-based generic model for describing tasks in ubiquitous computing environments. This model is used by a task selection algorithm that considers the context information of a user and the surrounding environment. Additionally, this paper proposes a pattern-based task reconfiguration algorithm. The algorithms are illustrated by a demo application conducted in our test bed, and by other examples of tasks selected under diverse situations. Evaluation results show a reasonable time overhead for the task selection algorithm.

Web data or data originated on the Web contain information and knowledge which allows to improve web site efficiency and effectiveness to attract and retain visitors.

However, web data have many irrelevant data inside. Consequently, it is necessary to preprocess them to model and understand the web user browsing behavior inside them. Further, due to frequent changes in the visitor’s behavior, as well as in the web site itself, the discovered knowledge may become obsolete in a short period of time.

In this paper, we introduce a platform which extracts, preprocesses and stores web data to enabling the utilization of web mining techniques. In other words, there is an Information Repository (IR) which stores preprocessed web data and it facilitates the patterns extraction. Likewise, there is a Knowledge Base (KB) for storing the discovered patterns which have been validated by a domain expert.

The proposed structure was tested using a real web site to prove the effectiveness of our approach.

We work on a Bayesian approach to the estimation of the specular component of a color image, based on the Dichromatic Reflection Model (DRM). The separation of diffuse and specular components is important for color image segmentation, to allow the segmentation algorithms to work on the best estimation of the reflectance of the scene. In this work we postulate a prior and likelihood energies that model the reflectance estimation process. Minimization of the posterior energy gives the desired reflectance estimation. The approach includes the illumination color normalization and the computation of a specular free image to test the pure diffuse reflection hypothesis.

Attribute Selection (AS) is generally applied as a data pre-processing step to sufficiently reduce the number of attributes in a dataset. This study uses six different data mining AS methods to identify a few key driving climate and air pollution attributes from small attribute sets (16 attributes) to increase knowledge about the underlying structures of acute respiratory hospital admission counts, because understanding key factors in environmental science data helps constructing a cost effective data collection and management process by focusing on collecting and investigating more representative and important variables. The performance of the selected attribute set was tested with Ant-Miner and C4.5 classifiers to examine the ability to prediction the admission count. Removal of attributes was successful over all AS methods, especially TNSU (a newly developed AS method, Tree Node Selection for unpruned), which achieved best in removing attributes and some improving the classification accuracy for Ant-Miner and C4.5. However, the overall prediction accuracy improvements are small, suggesting that AS selects attribute sets sufficiently enough to maintain the accuracy for Ant-Miner and C4.5.

In the paper a novel method of classification is presented. It is a combination of unsupervised and supervised techniques. First, the method divides the set of learning patterns into smaller ones in the clustering process. At the end of this phase a hierarchical structure of Self Organizing Map is obtained. Then for the leaves the classification rules are searched. To this end Bee Algorithm is used. The accuracy of the method was evaluated in an experimental way with the use of benchmark data sets and compared with the result of other methods.