Colin Fyfe – författare

The central idea of Hebbian Learning and Negative Feedback Networks is that artificial neural networks using negative feedback of activation can use simple Hebbian learning to self-organise so that they uncover interesting structures in data sets. Two variants are considered: the first uses a single stream of data to self-organise. By changing the learning rules for the network, it is shown how to perform Principal Component Analysis, Exploratory Projection Pursuit, Independent Component Analysis, Factor Analysis and a variety of topology preserving mappings for such data sets. The second variants use two input data streams on which they self-organise. In their basic form, these networks are shown to perform Canonical Correlation Analysis, the statistical technique which finds those filters onto which projections of the two data streams have greatest correlation. The book encompasses a wide range of real experiments and displays how the approaches it formulates can be applied to the analysis of real problems.

The central idea of Hebbian Learning and Negative Feedback Networks is that artificial neural networks using negative feedback of activation can use simple Hebbian learning to self-organise so that they uncover interesting structures in data sets. Two variants are considered: the first uses a single stream of data to self-organise. By changing the learning rules for the network, it is shown how to perform Principal Component Analysis, Exploratory Projection Pursuit, Independent Component Analysis, Factor Analysis and a variety of topology preserving mappings for such data sets. The second variants use two input data streams on which they self-organise. In their basic form, these networks are shown to perform Canonical Correlation Analysis, the statistical technique which finds those filters onto which projections of the two data streams have greatest correlation. The book encompasses a wide range of real experiments and displays how the approaches it formulates can be applied to the analysis of real problems.

Since its establishment in Hong Kong in 1998,the international Intelligent Data Engineering and Automated Learning (IDEAL) conference has become a r- erence for researchers in both theoretical and practical aspects of learning and information processing, data mining, retrieval and management, bioinformatics and bio-inspired models, agents and hybrid systems and ?nancial engineering. The purpose of IDEAL conferences has been to provide a broad and interd- ciplinary forum for scientists, researchers, and practitioners in these areas from around the world. A special feature of IDEAL conferences is cross-disciplinary exchange of ideas in emerging topics and application in these areas. Data ana- sis and engineering and associated learning paradigms are playing increasingly important roles in an increasing number of applications and ?elds. The mul- disciplinary nature of contemporary research is pushing the boundaries and one of the principal aims of the IDEAL conferences is to promote interactions and collaborations across disciplines. This volume of Lecture Notes in Computer Science contains accepted - pers presented at IDEAL 2006 held at the University of Burgos, Spain, during, September 20–23, 2006. The conference received 557 submissions from over 40 countriesaroundtheworld,whichweresubsequentlyrefereedbytheProgramme Committeeandmanyadditionalreviewers.Afterrigorousreview,170top-quality papers were accepted and included in the proceedings. The acceptance rate was only 30%, which ensured an extremely high-quality standard of the conference. The buoyant number of submitted papers is a clear proof of the vitality and increased importance of the ?elds related to IDEAL, and is also an indication of the rising popularity of the IDEAL conferences.

IDEAL 2008 was the ninth IDEAL conference to take place; earlier editions were held in Hong Kong, the UK, Australia and Spain. This was the first time, though hopefully not the last time, that it took place in Daejeon, South Korea, during November 2–5, 2008. As the name suggests, the conference attracts researchers who are involved in either data engineering or learning or, increasingly, both. The former topic involves such aspects as data mining (or intelligent knowledge discovery from databases), infor- tion retrieval systems, data warehousing, speech/image/video processing, and mul- media data analysis. There has been a traditional strand of data engineering at IDEAL conferences which has been based on financial data management such as fraud det- tion, portfolio analysis, prediction and so on. This has more recently been joined by a strand devoted to bioinformatics, particularly neuroinformatics and gene expression analysis. Learning is the other major topic for these conferences and this is addressed by - searchers in artificial neural networks, machine learning, evolutionary algorithms, artificial immune systems, ant algorithms, probabilistic modelling, fuzzy systems and agent modelling. The core of all these algorithms is adaptation.

Exploratory data analysis, also known as data mining or knowledge discovery from databases, is typically based on the optimisation of a specific function of a dataset. Such optimisation is often performed with gradient descent or variations thereof. In this book, we first lay the groundwork by reviewing some standard clustering algorithms and projection algorithms before presenting various non-standard criteria for clustering. The family of algorithms developed are shown to perform better than the standard clustering algorithms on a variety of datasets.We then consider extensions of the basic mappings which maintain some topology of the original data space. Finally we show how reinforcement learning can be used as a clustering mechanism before turning to projection methods. We show that several varieties of reinforcement learning may also be used to define optimal projections for example for principal component analysis, exploratory projection pursuit and canonical correlation analysis. The new method of cross entropy adaptation is then introduced and used as a means of optimising projections. Finally an artificial immune system is used to create optimal projections and combinations of these three methods are shown to outperform the individual methods of optimisation.

Theneedforintelligentsystemstechnologyinsolvingreal-lifeproblemshasbeen consistently growing. In order to address this need, researchers in the ?eld have been developing methodologies and tools to develop intelligent systems for so- ing complex problems. The International Society of Applied Intelligence (ISAI) through its annual IEA/AIE conferences provides a forum for international s- enti?c and industrial community in the ?eld of Applied Arti?cial Intelligence to interactively participate in developing intelligent systems, which are needed to solve twenty ?rst century's ever growing problems in almost every ?eld. The 23rdInternationalConference on Industrial, Engineering and Other - plications of Applied Intelligence Systems (IEA/AIE-2010) held in C' ordoba, Spain, followed IEA/AIE tradition of providing an international scienti?c forum for researchers in the ?eld of applied arti?cial intelligence. The presentations of theinvitedspeakersandauthorsmainlyfocusedondevelopingandstudyingnew methods to cope with the problems posed by real-life applications of arti?cial intelligence.Paperspresentedinthetwentythirdconferenceintheseriescovered theories as well as applications of intelligent systems in solving complex real-life problems.We received 297 papers for the main track, selecting 119 of them with the highest quality standards. Each paper was revised by at least three members of the Program Committee. The papers in the proceedings cover a wide number of topics including: applications to robotics, business and ?nancial markets, bio- formaticsandbiomedicine,applicationsofagent-basedsystems,computervision, control, simulation and modeling, data mining, decision support systems, evo- tionary computation and its applications, fuzzy systems and their applications, heuristic optimization methods and swarm intelligence, intelligent agent-based systems,internetapplications,knowledgemanagementandknowledgebaseds- tems, machine learning, neural network applications, optimization and heuristic search, and other real-life applications.

Theneedforintelligentsystemstechnologyinsolvingreal-lifeproblemshasbeen consistently growing. In order to address this need, researchers in the ?eld have been developing methodologies and tools to develop intelligent systems for so- ing complex problems. The International Society of Applied Intelligence (ISAI) through its annual IEA/AIE conferences provides a forum for international s- enti?c and industrial community in the ?eld of Applied Arti?cial Intelligence to interactively participate in developing intelligent systems, which are needed to solve twenty ?rst century's ever growing problems in almost every ?eld. The 23rdInternationalConference on Industrial, Engineering and Other - plications of Applied Intelligence Systems (IEA/AIE-2010) held in C' ordoba, Spain, followed IEA/AIE tradition of providing an international scienti?c forum for researchers in the ?eld of applied arti?cial intelligence. The presentations of theinvitedspeakersandauthorsmainlyfocusedondevelopingandstudyingnew methods to cope with the problems posed by real-life applications of arti?cial intelligence.Paperspresentedinthetwentythirdconferenceintheseriescovered theories as well as applications of intelligent systems in solving complex real-life problems.We received 297 papers for the main track, selecting 119 of them with the highest quality standards. Each paper was revised by at least three members of the Program Committee. The papers in the proceedings cover a wide number of topics including: applications to robotics, business and ?nancial markets, bio- formaticsandbiomedicine,applicationsofagent-basedsystems,computervision, control, simulation and modeling, data mining, decision support systems, evo- tionary computation and its applications, fuzzy systems and their applications, heuristic optimization methods and swarm intelligence, intelligent agent-based systems,internetapplications,knowledgemanagementandknowledgebaseds- tems, machine learning, neural network applications, optimization and heuristic search, and other real-life applications.

Theneedforintelligentsystemstechnologyinsolvingreal-lifeproblemshasbeen consistently growing. In order to address this need, researchers in the ?eld have been developing methodologies and tools to develop intelligent systems for so- ing complex problems. The International Society of Applied Intelligence (ISAI) through its annual IEA/AIE conferences provides a forum for international s- enti?c and industrial community in the ?eld of Applied Arti?cial Intelligence to interactively participate in developing intelligent systems, which are needed to solve twenty ?rst century's ever growing problems in almost every ?eld. The 23rdInternationalConference on Industrial, Engineering and Other - plications of Applied Intelligence Systems (IEA/AIE-2010) held in C' ordoba, Spain, followed IEA/AIE tradition of providing an international scienti?c forum for researchers in the ?eld of applied arti?cial intelligence. The presentations of theinvitedspeakersandauthorsmainlyfocusedondevelopingandstudyingnew methods to cope with the problems posed by real-life applications of arti?cial intelligence.Paperspresentedinthetwentythirdconferenceintheseriescovered theories as well as applications of intelligent systems in solving complex real-life problems.We received 297 papers for the main track, selecting 119 of them with the highest quality standards. Each paper was revised by at least three members of the Program Committee. The papers in the proceedings cover a wide number of topics including: applications to robotics, business and ?nancial markets, bio- formaticsandbiomedicine,applicationsofagent-basedsystems,computervision, control, simulation and modeling, data mining, decision support systems, evo- tionary computation and its applications, fuzzy systems and their applications, heuristic optimization methods and swarm intelligence, intelligent agent-based systems,internetapplications,knowledgemanagementandknowledgebaseds- tems, machine learning, neural network applications, optimization and heuristic search, and other real-life applications.

The IDEAL conference has become a unique, established and broad interdisciplinary forum for experts, researchers and practitioners in many fields to interact with each other and with leading academics and industries in the areas of machine learning, information processing, data mining, knowledge management, bio-informatics, neu- informatics, bio-inspired models, agents and distributed systems, and hybrid systems. This volume contains the papers presented at the 11th International Conference on Intelligent Data Engineering and Automated Learning (IDEAL 2010), which was held September 1–3, 2010 in the University of the West of Scotland, on its Paisley campus, 15 kilometres from the city of Glasgow, Scotland. All submissions were strictly pe- reviewed by the Programme Committee and only the papers judged with sufficient quality and novelty were accepted and included in the proceedings. The IDEAL conferences continue to evolve and this year’s conference was no exc- tion. The conference papers cover a wide variety of topics which can be classified by technique, aim or application. The techniques include evolutionary algorithms, artificial neural networks, association rules, probabilistic modelling, agent modelling, particle swarm optimization and kernel methods. The aims include regression, classification, clustering and generic data mining. The applications include biological information processing, text processing, physical systems control, video analysis and time series analysis.