By Topic

Intelligent Systems, 2008. IS '08. 4th International IEEE Conference

Date 6-8 Sept. 2008

Go

Filter Results

Displaying Results 1 - 25 of 26
  • Adaptive Fuzzy Clustering for improving classification performance in yeast data set

    Page(s): 18-2 - 18-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (367 KB) |  | HTML iconHTML  

    In data mining, there is inter-category imbalance of data which includes unnecessary data that hinder the formulation of an efficient model. This paper called FSFC+ introduces a new focused sampling based on adaptive fuzzy clustering. By applying FSFC+, the optimal number of clusters was used by adaptive method. It removes unuseful data that can be obstacles to the formulation of an efficient model. When there is no information about data set, we would evaluate the fitness of partitions produced by cluster validity index. In addition, it is very useful in data analysis because it can quantify the degree of membership of data to multiple clusters. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A hybrid method using PSO and NHL algorithms to train Fuzzy Cognitive Maps

    Page(s): 18-8 - 18-13
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (546 KB) |  | HTML iconHTML  

    In this paper a new hybrid method for training fuzzy cognitive maps is presented. FCMs are based on the knowledge of human experts and may not be accurate enough because of probable mistakes of experts. Thus, some learning methods have been investigated to train FCMs, so that these probable mistakes are covered. Two learning methods, PSO and NHL, and a new hybrid of them are introduced and implemented and tested for a chemical control problem. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Software for optimization of linear objective function with fuzzy relational constraint

    Page(s): 18-14 - 18-19
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (451 KB) |  | HTML iconHTML  

    The paper presents method, algorithm and software in MATLAB and Java for solving various optimization problems when the linear objective function is subject to fuzzy linear system of equations or fuzzy linear systems of inequalities as constraint. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fuzzy Neural Network for detecting nonlinear determinism in gastric electrical activity: Fractal dimension approach

    Page(s): 18-20 - 18-23
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (474 KB) |  | HTML iconHTML  

    A robust method of detecting determinism for short time series is proposed and applied to both healthy and Functional Gastrointestinal Activity of GEA signals. The method provides a robust measure of determinism through characterizing the trajectories of the signal components which are obtained through singular value decomposition. In order to automatically differentiate the gastric function, a fuzzy neural network to classify the types based on the knowledge of qualified knowledge in chaotic features differences in diagnosis was designed. The designed classifier can make hard decision and soft decision for identifying the chaotic patterns of GMA signal at the accuracy of 91%, which is better than the results that achieved by back-propagation neural network. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • The development of a hybrid, distributed architecture for multiagent systems and its application in robot soccer

    Page(s): 19-2 - 19-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (700 KB) |  | HTML iconHTML  

    Several issues still need to be unraveled in the development of multiagent systems equipped with global vision, as in robot soccer leagues. Here, we underscore three of them (1) real-time constraints on recognition of scene objects; (2) acquisition of environment knowledge; and (3) distribution and allocation of control competencies shared between the repertoire of the agentpsilas reactive behavior, and the central control entitypsilas strategic and deliberative behavior. The objective of this article is to describe the implementation of a distributed and hybrid reactive-deliberative control architecture for a multiagent system, equipped with global vision camera and agent local sensor and cameras. This multiple agent system was developed for application in robot soccer. We present the digital image processing techniques applied, as well as the proposed control architecture aimed at satisfying the constraints of this kind of application. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A relational learning approach to activity recognition from sensor readings

    Page(s): 19-8 - 19-13
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (851 KB) |  | HTML iconHTML  

    The ability of understanding humanpsilas behavior is a required component for many applications. This understanding includes, among other tasks, automatically generating and maintaining models of human actions, goals and plans. This paper presents a system to infer the actions that people perform in order to accomplish activities of daily living starting from sensory inputs. Our approach is based on using relational learning to infer predictions about which action has just been executed. We learn a model for recognizing executed actions based on the state changes detected from sensor readings. Each change has been produced by a performed action, while a sequence of these actions forms a plan to accomplish a high-level action or to achieve a goal. Using a relational learning tool, Tilde, we obtain classifiers to map changes in the states to actions performed by a user. We have performed some experiments using an environment simulator feeded by data gathered from real human behaviour. The results show that we can obtain a good accuracy even in presence of noise. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Utilization of K-NN algorithm for expectation maximization based classification method

    Page(s): 19-14 - 19-16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (634 KB) |  | HTML iconHTML  

    K nearest neighbor and Bayesian algorithms are effective methods of machine learning. In this work a data elimination approach is proposed to improve data clustering. The proposed method is based on hybridization of K nearest neighbor and Bayesian learning algorithms. The suggested method is tested on well-known machine learning data sets such as iris, wine and breast cancer and the results are concluded. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Supply chain optimization towards personalizing web services

    Page(s): 19-17 - 19-22
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (465 KB) |  | HTML iconHTML  

    Personalization, which has the ultimate goal of satisfying userpsilas requests, can be perceived in terms of QoS measurement. As one of the means for the success of semantics Web, many techniques have been effectively used in modeling and developing Web service personalization. However, most of these methodologies relied heavily on detailed implicit and explicit information supply by users during initial and subsequent interactions with the systems. We propose in this paper a novel approach using the supply chain management (SCM) technique in personalizing Web services as against the conventional notion of applying SCM only to product manufacturing. Our user-model based framework uses multi-agent system (MAS) components in taking requests from users and working towards their satisfaction including seeking for additional information outside the system as the need arises. Only basic stereotype information furnished by potential users at initial contact is required for personalization during subsequent interactions with the system. The system is adaptive and aimed at high quality autonomous information services where users are successfully presented preferred Web services with minimum information request. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A combined genetic algorithm and Sugeno fuzzy logic based approach for on-line tuning in pH process

    Page(s): 20-2 - 20-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (385 KB) |  | HTML iconHTML  

    Computing the optimal values of proportional integral derivative (PID) control gains is an important task in the design of PID controller. This paper presents the application of Sugeno fuzzy model for on-line tuning of PID controllers in pH process. The optimal PID controller parameters required to develop the Sugeno fuzzy model are estimated by genetic algorithm. The developed fuzzy controller can give the PID parameters on line for different operating conditions. The suitability of the proposed approach has been demonstrated through computer simulation using MATLAB Simulink. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Liquidity management using Cash Flow at Risk

    Page(s): 20-8 - 20-12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (453 KB) |  | HTML iconHTML  

    The article describes the methodology and practical implementation of Cash Flow at Risk (CFaR) calculation with the help of a multidimensional Monte Carlo simulation. CFar is one of the most important characteristics of liquidity. It assesses the likelihood that operating cash flows will drop below a prespecified level. In this paper is considered application for measuring CFaR under impact of different environment factors. The application is written in C# .NET. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Volterra model predictive control of a lyophilization plant

    Page(s): 20-13 - 20-18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (730 KB) |  | HTML iconHTML  

    Lyophilization plants are widely used by pharmaceutical industries to produce stable dried medications and important preparations. Since, a Lyophilization cycle involves a high energy demands it is needed to be used an improved control strategy in order to minimize the operating costs. This paper describes a method for designing a nonlinear model predictive controller to be used in a Lyophilization plant. The controller is based on a truncated fuzzy-neural Volterra predictive model and a simplified gradient optimization algorithm. The proposed approach is studied to control the product temperature in a Lyophilization plant. The efficiency of the proposed approach is tested and proved by simulation experiments. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Application of the Analytic Network Process (ANP) in a framework of ERP systems implementation success

    Page(s): 20-19 - 20-24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1214 KB) |  | HTML iconHTML  

    The objective of this paper is to propose a framework of ERP systems implementation success that helps project managers to deal with large and complex ERP systems implementation projects. The framework is designed to evaluate and to select activities or alternatives to improvement the degree of performance by the success factors by using the analytic network process (ANP) and the priority matrix (PM). The experience with the application of ANP in a framework of ERP systems implementation success suggests that it offers guidelines in the evaluation and selection of alternatives. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A discernibility-based approach to feature selection for microarray data

    Page(s): 21-2 - 21-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (475 KB) |  | HTML iconHTML  

    Feature selection has been used widely for a variety of data, yielding higher speeds and reduced computational cost for the classification process. However, it is in microarray datasets where its advantages become more evident and are more required. In this paper we present a novel approach to accomplish this based on the concept of discernibility that we introduce to depict how separated the classes of a dataset are. We develop and test two independent feature selection methods that follow this approach. The results of our experiments on four microarray datasets show that discernibility-based feature selection reduces the dimensionality of the datasets involved without compromising the performance of the classifiers. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Neural network application in strange attractor investigation to detect a FGD

    Page(s): 21-8 - 21-11
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (482 KB) |  | HTML iconHTML  

    There is growing interest in modeling and processing nonlinear behavior in the biological systems. In this paper we applied such methods for detecting Functional Disorder in Gastric. Conventional tools for analyzing such data use information from the power spectral density of the time series, and hence are restricted to little information of data. This information does not provide a sufficient representation of a signal with strong nonlinear properties. In this work, we attempt to extract various nonlinear dynamical invariants of the underlying attractor from the signals. We show that these invariants can discriminate between normal and Functional Gastrointestinal Disorders (FGD) classes. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Transforming relational model to source ontology for data integration

    Page(s): 22-2 - 22-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (501 KB) |  | HTML iconHTML  

    Data integration can be made scalable, flexible and interoperable if the source descriptions are represented in conceptual model i.e., ontology. Ontology is a shared and common understanding of a domain and can solve the heterogeneity of distributed data sources. The methodology provided in this paper transforms database relations of a local source to OWL based ontology for source descriptions. It minimizes the effort and errors involved in manual ontology building. Compared with existing techniques the distinguished feature of the proposed technique is to build ontology in the absence of necessary metadata from physical or logical models. Results of the proposed methodology are provided to show the transformation is correct (i.e., total, injective). View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ontology-driven relevance reasoning architecture for data integration techniques

    Page(s): 22-8 - 22-13
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (564 KB) |  | HTML iconHTML  

    In order to execute a userpsilas query in a data integration system, the query execution process needs to be optimized. Before executing a query at real time, relevant and effective data sources must be identified. In this paper we propose an ontology-driven relevance reasoning architecture for future data integration techniques that will improve the response time for queries during the relevance reasoning process. Ontology has played a vital role to develop various component of the architecture. Source descriptions are plotted over the bitmap index in an intelligent and improved manner. Despite taking a lot of time in traversing local ontologies of source descriptions, bitmap index is exploited in relevance reasoning to identify the relevant and most effective data sources for userpsilas query. These identified data sources are ranked based on their relevance to the userpsilas query and then queried accordingly. A distinguished feature of the system is that it facilitates the user to write the query in terms of their local ontology concepts as well as global ontology concepts. A brief discussion is done on the results of the experimental study of proposed methodology for relevance reasoning and improvements are shown as compared to the previous systems. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ontology based semantic information retrieval

    Page(s): 22-14 - 22-19
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (466 KB) |  | HTML iconHTML  

    Semantic-based information retrieval techniques understand the meanings of the concepts that users specify in their queries. The main drawback of the existing semantic-based information retrieval techniques is that none of them considers the context of the concept(s). We propose a semantic information retrieval framework to improve the precision of search results. In this paper, thematic similarity approach is employed for information retrieval in order to capture the context of particular concept(s). We store metadata information of source(s) in the form of RDF triples. We search userpsilas queries in the existing metadata by matching RDF triples instead of keywords. The results of the experiments performed on our framework showed improvements in precision and recall compared to the existing semantic-based information retrieval techniques. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • System architecture for efficient Grid Resource Management

    Page(s): 22-20 - 22-25
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (496 KB) |  | HTML iconHTML  

    Efficient management of continuously increasing resources is a major challenging task in grid research community. Further, efficient scheduling - an integral part of the management process - requires proficient resource organization as evidenced in available literature in this research domain. Some potential resources to schedule a given job are not discovered because existing systems are based on syntactic resource matching. However they can partially fulfill the request by different degrees of match. Semantic metadata and ontologies are considered helpful in managing resources more effectively. In this paper we investigate and explore how grid resource management can benefit from semantic knowledge. We propose semantic-based grid resource management (S-GRM) system which utilizes semantic metadata to describe and discover both logical and physical resources .We also discussed a prototype implementation of our proposed architecture using ProActive grid middleware. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • PES: Personalization and evaluation system based on multi-agents approach: Application in transport information

    Page(s): 23-2 - 23-6
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2534 KB) |  | HTML iconHTML  

    In recent years, personalized information systems (PIS), and mainly in the field of transport, are increasingly developing. In fact, their aim is to provide the user with the relevant information that directly interests him/her and suits his/her preferences. There exist several personalization methods yet, to our knowledge and at the present point of research, the evaluation of such systems was ignored. To fill up this lack, it is necessary to envisage new personalization methods taking into account the evaluation function in order to guarantee a good quality of services that these systems may provide. In this paper, we study some existing PIS. Then, we propose a personalization and evaluation method based on multi-agent approach. Finally, we present some experimentation results in transport field to validate the mentioned approach. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Ontology based semantic interoperability facilitator among task group

    Page(s): 23-7 - 23-10
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1491 KB) |  | HTML iconHTML  

    Collaboration among enterprises in a dynamic environment makes the actors to concentrate on their respective core competences and allow provision and sharing of expertise, resources, and skills for taking advantages and better respond to business opportunities. The coming together of such organizations, usually enhanced by computer network, is referred to as virtual enterprise (VE) or task group. This partnership is only possible if the systems in the various associated organizations can process the data in one another. A major challenge in the enterprise collaborative system that has attracted many research efforts in the recent past is semantic interoperability. This collaborative-based market place requires among others common conceptualization and meaningful data exchange. Effective collaboration among VE members is a major key to the accomplishment of this noble objective. For this interoperability to be effective, we propose in this paper an ontology-based middleware framework ldquoOntology Gatewayrdquo used by players in such situation to exchange information needed to carry out the process. The middleware not only assists in the formation of VE by interested members, but also facilitates semantic interpretability among task groups. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Genetic Algorithm approach for Optimal Power Flow with FACTS devices

    Page(s): 23-11 - 23-16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (434 KB) |  | HTML iconHTML  

    This paper presents a genetic algorithm (GA) based approach for solving the optimal power flow with FACTS devices to eliminate line over loads in the system following single line outages. The proposed approach introduces an index called the single contingency sensitivity (SCS) index to rank the system branches according to their suitability for installing thyristor controlled series capacitors(TCSCs). Once the locations are determined, the problem of identifying the optimal TCSC parameters is formulated as an optimization problem and a GA based approach is applied to solve the optimal power flow (OPF) problem. IEEE 30 bus system is considered to demonstrate the suitability of this algorithm. Case studies on IEEE test system show the effectiveness of this algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Design of a context script language for developing context-aware applications in ubiquitous intelligent environment

    Page(s): 23-17 - 23-23
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (769 KB) |  | HTML iconHTML  

    In this paper, we design an overall architecture for dealing with context-aware applications in ubiquitous intelligent environment. The architecture is composed of middleware, context server, and client. The middleware plays an important role in recognizing a moving node with mobility by using a Bluetooth wireless communication technology as well as in executing an appropriate execution module according to the context acquired from a context server. To develop context-aware applications in an effective manner under this architecture, we propose a new context script language which can be used to represent both various decisions on context-awareness and appropriate procedures according to the decision as a standard syntax. Thus, it is regarded as a general purpose language to provide users with functions to define a given context in a clear and precise way. We also design and implement the processor of our context script language, which can automatically execute a series of involved procedures acquired for context-awareness. To show the usefulness of our context script language, we develop a context-aware application which can provide users with a music playing service in ubiquitous intelligent environment. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A case-based planning approach to design and plan ITMAS

    Page(s): 23-24 - 23-30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1644 KB) |  | HTML iconHTML  

    Case-based planning which is an induction-based mechanism, has many practical applications in designing and planning of multi-agent systems, which has been recently known as a powerful processing tool in transactional and interactive environments. Intelligent tutoring systems, that have drawn the most attention in the computational mechanisms of concept transition in recent years, have successfully put into practice the concept of multi-agent interaction as well. This paper describes the advances applied in our intelligent tutoring multi-agent system architecture which facilitate meeting the requirements of intelligent tutoring systems and their users. In addition throughout this paper it is explained how case-based planning is utilized for designing and planning the proposed intelligent tutoring multi-agent system to attain an optimized final plan for current user or environment. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Methodology and tools for the design and verification of a smart management system for home comfort

    Page(s): 24-2 - 24-7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (887 KB) |  | HTML iconHTML  

    Measurement, control and monitoring systems are increasingly complex and may require various disciplinary skills. Producing such systems requires rigorous design methodology and the right tools. It is with this in mind that the design and formal verification HiLes tool was developed. This methodological design and verification process was used in this work to design a complex system dedicated to the intelligent home comfort management. It is built into the system engineering process defined by EIA-632, which uses UML and SYSML standards to identify the requirements, model the control logics of the system and formally validate its dynamic properties, all in a semi-automated way. In this paper, we present the various design phases according to this method as well as the results of formal verification of the complete system dynamic behavior. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-organizing maps for automatic fault detection in a vehicle cooling system

    Page(s): 24-8 - 24-12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (491 KB) |  | HTML iconHTML  

    A telematic based system for enabling automatic fault detection of a population of vehicles is proposed. To avoid sending huge amounts of data over the telematics gateway, the idea is to use low-dimensional representations of sensor values in sub-systems in a vehicle. These low-dimensional representations are then compared between similar systems in a fleet. If a representation in a vehicle is found to deviate from the group of systems in the fleet, then the vehicle is labeled for diagnostics for that subsystem. The idea is demonstrated on the engine coolant system and it is shown how this self-organizing approach can detect varying levels of clogged radiator. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.