By Topic

Intelligent Systems (IS), 2010 5th IEEE International Conference

Date 7-9 July 2010

Filter Results

Displaying Results 1 - 25 of 89
  • [Front cover]

    Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (1578 KB)  
    Freely Available from IEEE
  • [Copyright notice]

    Page(s): ii
    Save to Project icon | Request Permissions | PDF file iconPDF (32 KB)  
    Freely Available from IEEE
  • Table of contens

    Page(s): iii - xiv
    Save to Project icon | Request Permissions | PDF file iconPDF (753 KB)  
    Freely Available from IEEE
  • Message from the conference chairs

    Page(s): xv - xvii
    Save to Project icon | Request Permissions | PDF file iconPDF (65 KB)  
    Freely Available from IEEE
  • Self-configurable framework for enabling context-aware learning design

    Page(s): 1 - 6
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (770 KB) |  | HTML iconHTML  

    There have been attempts to offer “toolkits” or software to enable ease of entry into pedagogic design and support non-specialists in engaging with learning theories. Despite the efforts, existing e-learning systems and authoring tools have several limitations in respect of support provided and cannot accommodate the needs of teachers who increasingly look for more intelligent services and support when designing. Drawing not just upon the semantic web but also knowledge management and autonomic computing we investigate how to design and build next generation learning design environments that support the necessary context-aware features. The design of a self-configurable framework that uses semantic web core ideas and fundamental knowledge-management inferencing is presented. We conclude with some initial findings. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Bibliography mapping with semantic social bookmarks

    Page(s): 7 - 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1501 KB) |  | HTML iconHTML  

    A semantically-rich and structured bibliography can be seen as a map of the publication space related to the research area. Social bookmarking applications have shown considerable advantage over traditional reference applications in creating and managing such “publication maps” by means of personalisation, recommendation and global availability. Advances in semantic web and linked data are increasing the need for means of guided semantic navigation that the existing bookmarking systems lack. As a solution, we consider a model of semantic bookmarking that is personalised, semantically rich and navigationally savvy. The emphasis is made on separating structural and semantic layers in order to improve navigation by maintaining and visualising structural relationships among the bookmarks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Identifying user strategies in exploratory learning with evolving task modelling

    Page(s): 13 - 18
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (944 KB) |  | HTML iconHTML  

    In this paper we present work on adaptive identification of learners' strategies, gradually developing a higher level of adaptation based on evolving models of mathematical generalisation tasks in an Exploratory Learning Environment. A similarity-based classification approach is defined for the identification of strategies, using an initially small number of classes (i.e. strategies). A strategy is composed of several patterns with relations between them. An evolution monitor component observes changes in the environment and triggers a mechanism that builds-up the task model. The task model evolves when new relevant information becomes available by adding a new strategy (class) or a new inefficient pattern, i.e. patterns that make it difficult for the learner to generalise. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A generic framework for enhancing the interpretability Of granular computing-based information

    Page(s): 19 - 24
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1380 KB) |  | HTML iconHTML  

    One of the main advantages of Granular Computing and Fuzzy Logic is the transparency and interpretability features that are available to the user. In this paper we present a systematic data granulation algorithm for the elicitation of Fuzzy rules and show how the granular data and relational information extracted during the data mining process can be translated into Fuzzy Logic statements with enhanced interpretability. Notions of granular cardinality, distribution and distance are used to apply linguistic hedges to two-sided Gaussian Fuzzy membership functions. The proposed methodology is applied to a biomedical dataset relating to Electrical Impedance Tomography (EIT) measurements of lung ventilation showing good agreement and interpretability between the captured knowledge and the theoretical and physiological expectations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Enhancing functionality of complex plant hybrid control system using case-based reasoning

    Page(s): 25 - 30
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (176 KB) |  | HTML iconHTML  

    A hybrid control system with Case Based Reasoning (CBR) procedure is considered. It is intended for complex industrial plants in abnormal and fault situations where the conventional control systems cannot work autonomously. CBR systems allow capture and use a specific experience by Case Base (CB) creation, similarity assessment, case adaptation and learning new cases. Some new results are incorporated into the traditional CBR cycle, relevant to control tasks - uncertainty, presence of multivariant solutions from different operators, rational distribution of knowledge among autonomous agents, ontologies and CBR-system. Simulation results are represented for an industrial plant with a non-square matrix. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • ‘Symbiotic’ data-driven modelling for the accurate prediction of mechanical properties of alloy steels

    Page(s): 31 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (855 KB) |  | HTML iconHTML  

    A new optimal strategy based on symbiotic modelling is proposed. The system combines Linear Regression Model (LR), Non-Linear Iterative Partial Adaptive Least Square Model (NIPALS), Neural Network Model with double loop procedures (NNDLP), Adaptive Numeric Modelling (Neural-Fuzzy modeling NF) and metallurgical knowledge in order to provide effective modelling solutions and achieve an optimal prediction performance. As a final step a fusion procedure is used to perform a routine decision making based on aggregation algorithm and clustering method that allow to systematically select the final best prediction outcome from a set of competing solutions. The proposed methodology is then applied to the challenging environment of a multi-dimensional, non-linear and sparse data space consisting of mechanical properties of `Mild' Steel in particular Tensile Strength (TS) and Yield Strength (YS) in hot-rolling industrial processes. Using a data set containing critical information on the mechanical properties obtained from a hot strip mill, it is concluded that the developed new systematic modelling approach is capable of providing better prediction than each individual model even in data distribution areas which are reckoned to be sparse. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Adaptive PI type iterative learning control

    Page(s): 37 - 42
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (653 KB) |  | HTML iconHTML  

    Based on combination of an optimal PI type iterative learning controller and projection like adjusting algorithm, an adaptive iterative learning control scheme is presented for repetitive control of uncertain systems. This adaptive iterative learning controller is designed without any priori knowledge of system parameters. The convergence of the presented scheme is analyzed and its convergence condition is obtained in terms of adjusting algorithm step size. An illustrative example is given to demonstrate the effectiveness of the proposed technique. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Inertial Navigation aided by Simultaneous Localization and Mapping

    Page(s): 43 - 48
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (957 KB) |  | HTML iconHTML  

    Inertial and vision sensors are must nowadays in terms of navigation and guidance measurements for autonomous aerial and ground vehicles. To make up for the Inertial Navigation divergence the concept of aiding Inertial Navigation by Simultaneous Localization and Mapping is introduced. In this paper we describe the needing changes to the Simultaneous Localization and Mapping augmented state vector and show that repeated measurements of a map point with certain maneuvers around or by the map point are crucial for constraining the Inertial Navigation position divergence and reducing the covariance of the map points. This integration makes a passive jamproof GPS free autonomous navigation system. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Attitude determination from single camera vector observations

    Page(s): 49 - 54
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (764 KB) |  | HTML iconHTML  

    Attitude determination is of major importance in Guidance and Control Systems of the Unmanned Aerial Vehicles (UAV's). Supplying wrong or not precise attitude very often turns to be catastrophic for the UAV's. Vision sensors are must nowadays. They provide reach source of information given as relative measurements between the vehicle navigation parameters (position, velocity and attitude) and the environment. This paper presents a framework for attitude determination from single camera vector observations. We assume known environment in a form of a map and true vehicle positions from which each observation has been taken. Two different methods for attitude determination are presented: an iterative numerical solution based on Gauss Newton's method and an exact method known as the Davenport q-method. Pros and cons of the both solutions are presented. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A decomposition-based approach to flexible flow shop scheduling under stochastic setup times

    Page(s): 55 - 60
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (757 KB) |  | HTML iconHTML  

    Research on production scheduling under uncertainty has recently received much attention. This paper presents a novel decomposition-based approach (DBA) to flexible flow shop (FFS) scheduling under stochastic setup times. In comparison with traditional methods using a single approach, the proposed DBA combines and takes advantage of two different approaches, namely the Genetic Algorithm (GA) and the Shortest Processing Time Algorithm (SPT), to deal with uncertainty. A neighbouring K-means clustering algorithm is developed to firstly decompose an FFS into an appropriate number of machine clusters. A back propagation network (BPN) is then adopted to assign either GA or SPT to generate a sub-schedule for each machine cluster. Finally, an overall schedule is generated by integrating the sub-schedules of the machine clusters. Computation results reveal that the DBA is superior to SPT and GA alone for FFS scheduling under stochastic setup times. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Collaborative ontology building using qualitative information collection methods

    Page(s): 61 - 66
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (263 KB) |  | HTML iconHTML  

    In the actual competitive context, doing business globally has become critical to the survival of most enterprises. To achieve it, enterprises require the establishment of cooperation agreements among each other. Thus, there is a demand for intelligent solutions capable of reinforcing partnerships and collaborations. However, due to the worldwide diversity of communities, a high number of knowledge representation elements, as ontologies, which are not semantically coincident, have appeared representing the same segment of reality. Therefore, even in the same domain, enterprises do not understand each other, impeding various systems parties to seamless communicate. To solve this semantic interoperability problem, it has been suggested to build a reference ontology able to represent such cluster of interoperating entities. The authors propose a collaborative ontology building methodology, enriched with qualitative information collection methods, to effectively improve the approach to elicit knowledge from business domain experts, towards interoperable intelligent systems. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Self-learning strategies in multiagent environments

    Page(s): 67 - 72
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (102 KB) |  | HTML iconHTML  

    A method for revealing and resolving conflicts is presented, especially well applicable for resolution of contradictions. It is shown that agents receive a greater autonomy via a correctly directed or selectable identification and conflict resolution inside the accumulated knowledge. The advantages of the introduced method are presented compared to artificial neural networks (ANN) and other trainable tools with or without a teacher. The offered material is tightly bound to the research presented at former conferences IS'02 and IS'08. Applications are oriented mainly to serve the goals of information security but for the sake of brevity their descriptions cross the borders of the present paper. The introduced method is no less successive in applications for abstract and applied mathematics, in neuro-fuzzy systems, etc. The method may be efficiently combined with other well-known methods and technologies: ANN, machine learning, statistical learning and data mining, knowledge discovery etc. In case of combined exploitation with ANNs a `critical learner' may be constructed who should establish an active feedback with the teacher and make a deep learning. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An intelligent system for semi-automatic evolution of ontologies

    Page(s): 73 - 78
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (685 KB) |  | HTML iconHTML  

    Ontologies are an important part of the Semantic Web as well as of many intelligent systems. However, the traditional expert-driven development of ontologies is time-consuming and often results in incomplete and inappropriate ontologies. In addition, since ontology evolution is not controlled by end users, it may take too long for a conceptual change in the domain to be reflected in the ontology. In this paper, we present a recommendation algorithm in a Web 2.0 platform that supports end users to collaboratively evolve ontologies by suggesting semantic relations between new and existing concepts. We use the Wikipedia category hierarchy to evaluate our algorithm and our experimental results show that the proposed algorithm produces high quality recommendations. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Towards the development of genuine intelligent ontology-based e-Learning systems

    Page(s): 79 - 84
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (760 KB) |  | HTML iconHTML  

    Intelligence of present e-Learning systems is usually static: these systems may provide learning materials of different complexity and/or in differing sequence to various learners according to their abilities, skills and learning progress, but these systems hardly themselves have the ability to evolve and learn new knowledge during their life-cycle. We consider intelligence, which is characterized by (semi-)autonomous knowledge acquisition, learning and/or reasoning in order to enable the provision of better services to the users, as a system quality attribute. We propose a framework for the extension of an existing e-Learning system with intelligence capabilities. We describe an intelligent component of an e-Learning system that is capable of enriching its local domain ontology with new concepts and relationships obtained by querying a remote knowledge base. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Storage, degradation and recall of agent memory in Serious Games and Simulations

    Page(s): 85 - 90
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1163 KB) |  | HTML iconHTML  

    This paper describes our novel method for creating AI Agents that store and recall memories in a biologically inspired manner. We consider research on biological systems and note the limits on these systems in terms of both storage and recall capacity. We also examine existing technological methods for storing memories in Intelligent Agent systems and discuss their limitations for use in Serious Games and Simulations. We then suggest our own method for storing memories, degrading them over time, arranging and searching them and retrieving them in a realistic fashion. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Agent-oriented middleware for InfoStation-based mLearning intelligent systems

    Page(s): 91 - 95
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (720 KB) |  | HTML iconHTML  

    An agent-oriented middleware supporting contextaware and adaptable mLearning service provision within an InfoStation-based University network is presented. The InfoStation's middleware architecture facilitating the users' mobile (WiFi) access to services is described. The agents' interaction is explained in detail. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Estimating the relevance of a data source using a fuzzy-cardinality-based summary

    Page(s): 96 - 101
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (611 KB) |  | HTML iconHTML  

    In this paper, we consider the situation where a fuzzy query is submitted to distributed data sources. In order to save bandwith and processing cost, we propose a technique whose aim is to forward the query to the most relevant sources only. It is assumed that a fuzzy summary of every data source is available, and the approach we propose consists in estimating the relevance of a source wrt to a user query, based on its associated summary. The general case where the user does not necessarily employ the vocabulary (i.e., the labels from the fuzzy partitions) that was used for summarizing the source is considered. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Support driven opportunistic aggregation for generalized itemset extraction

    Page(s): 102 - 107
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (672 KB) |  | HTML iconHTML  

    Association rule extraction is a widely used exploratory technique which has been exploited in different contexts (e.g., biological data, medical images). However, association rule extraction, driven by support and confidence constraints, entails (i) generating a huge number of rules which are difficult to analyze, or (ii) pruning rare itemsets, even if their hidden knowledge might be relevant. To address the above issues, this paper presents a novel frequent itemset mining algorithm, called GENIO (GENeralized Itemset DiscOverer), to analyze correlation among data by means of generalized itemsets, which provide a powerful tool to efficiently extract hidden knowledge, discarded by previous approaches. The proposed technique exploits a (user provided) taxonomy to drive the pruning phase of the extraction process. Instead of extracting itemsets for all levels of the taxonomy and post-pruning them, the GenIO algorithm performs a support driven opportunistic aggregation of itemsets. Generalized itemsets are extracted only if itemsets at a lower level in the taxonomy are below the support threshold. Experiments performed in the network traffic domain show the efficiency and the effectiveness of the proposed algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Array-Tree: A persistent data structure to compactly store frequent itemsets

    Page(s): 108 - 113
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (580 KB) |  | HTML iconHTML  

    Frequent itemset mining discovers correlations among data items in a transactional dataset. A huge amount of itemsets is often extracted, which is usually hard to process and analyze. The efficient management of the extracted frequent itemsets is still an open research issue. This paper presents a new persistent structure, the Array-Tree, that compactly stores frequent itemsets. It is an array-based structure exploiting both prefix-path sharing and subtree sharing to reduce data replication in the tree, thus increasing its compactness. The Array-Tree can be profitably exploited to efficiently query extracted itemsets by enforcing user-defined item or support constraints. Experiments performed on real and synthetic datasets show both the compactness of the Array-Tree data representation and its efficient support to user queries. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • DB-SMoT: A direction-based spatio-temporal clustering method

    Page(s): 114 - 119
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (931 KB) |  | HTML iconHTML  

    Existing works for semantic trajectory data analysis have focused on the intersection of trajectories with application important geographic information and the use of the speed to find interesting places. In this paper we present a novel approach to find interesting places in trajectories, considering the variation of the direction as the main aspect. The proposed approach has been validated with real trajectory data associated to oceanic fishing vessels, with the objective to automatically find the real places where vessels develop fishing activities. Results have demonstrated that the method is very appropriate for applications in which the direction variation plays the essential role. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Detection and tracking of multiple moving objects with occlusion in smart video surveillance systems

    Page(s): 120 - 125
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (944 KB) |  | HTML iconHTML  

    Autonomous video surveillance and monitoring has a rich history. A new method for detecting and tracking multiple moving objects based on discrete wavelet transform and identifying the moving objects by their color and spatial information is proposed in this paper. Since discrete wavelet transform has a nice property that it can divide a frame into four different frequency bands without loss of the spatial information, it is adopted to solve this problem due to the fact that most of the fake motions in the background can be decomposed into the high frequency wavelet sub-band. In tracking multiple moving objects, many applications have problems when objects pass across each other. In this paper, we have developed robust routines for detecting and tracking multiple moving objects with occlusion. The proposed model has proved to be robust in various environments (including indoor and outdoor scenes) and different types of background scenes. The experimental results prove the feasibility of the proposed method. Experiments on real scenes show that the algorithm is effective for object detection and tracking. View full abstract»

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