Classification

This book explains the minimum error entropy (MEE) concept applied to data classification machines. Theoretical results on the inner workings of the MEE concept, in its application to solving a variety of classification problems, are presented in the wider realm of risk functionals.

Foundations of Computational Intelligence Volume 5: Function Approximation and Classification Approximation theory is that area of analysis which is concerned with the ability to approximate functions by simpler and more easily calculated functions. It is an area which, like many other fields of analysis, has its primary roots in the mat- matics. The need for function approximation and classification arises in many branches of applied mathematics, computer science and data mining in particular. This edited volume comprises of 14 chapters, including several overview Ch- ters, which provides an up-to-date and state-of-the art research covering the theory and algorithms of function approximation and classification. Besides research ar- cles and expository papers on theory and algorithms of function approximation and classification, papers on numerical experiments and real world applications were also encouraged.

The present volume contains a selection of papers presented at the Eighth Conference of the International Federation of Classification Societies (IFCS) which was held in Cracow, Poland, July 16-19, 2002. All originally submitted papers were subject to a reviewing process by two independent referees, a procedure which resulted in the selection of the 53 articles presented in this volume. These articles relate to theoretical investigations as well as to practical applications and cover a wide range of topics in the broad domain of classifi­ cation, data analysis and related methods. If we try to classify the wealth of problems, methods and approaches into some representative (partially over­ lapping) groups, we find in particular the following areas: • Clustering • Cluster validation • Discrimination • Multivariate data analysis • Statistical methods • Symbolic data analysis • Consensus trees and phylogeny • Regression trees • Neural networks and genetic algorithms • Applications in economics, medicine, biology, and psychology.

Software similarity and classification is an emerging topic with wide applications. It is applicable to the areas of malware detection, software theft detection, plagiarism detection, and software clone detection. Extracting program features, processing those features into suitable representations, and constructing distance metrics to define similarity and dissimilarity are the key methods to identify software variants, clones, derivatives, and classes of software.

The growing capabilities in generating and collecting data has risen an urgent need of new techniques and tools in order to analyze, classify and summarize statistical information, as well as to discover and characterize trends, and to automatically bag anomalies. This volume provides the latest advances in data analysis methods for multidimensional data which can present a complex structure: The book offers a selection of papers presented at the first Joint Meeting of the Société Francophone de Classification and the Classification and Data Analysis Group of the Italian Statistical Society. Special attention is paid to new methodological contributions from both the theoretical and the applicative point of views, in the fields of Clustering, Classification, Time Series Analysis, Multidimensional Data Analysis, Knowledge Discovery from Large Datasets, Spatial Statistics.

This book introduces a fuzzy classification approach, which combines relational databases with fuzzy logic for more effective and powerful customer relationship management (CRM). It shows the benefits of a fuzzy classification in contrast to the traditional sharp evaluation of customers for the acquisition, retention and recovery of customers in online shops. The book starts with a presentation of the basic concepts, fuzzy set theory, and the combination of relational databases and fuzzy classification. In its second part, it focuses on the customer perspective, detailing the central concepts of CRM, its theoretical constructs and aspects of analytical, operational and collaborative CRM.

Digital libraries (DLs) have introduced new technologies, as well as leveraging, enhancing, and integrating related technologies, since the early 1990s. These efforts have been enriched through a formal approach, e.g., the 5S (Societies, Scenarios, Spaces, Structures, Streams) framework, which is discussed in two earlier volumes in this series. This volume should help advance work not only in DLs, but also in the WWW and other information systems.