Synopses & Reviews
Classification & data characteristics are important issues that can be found in many disciplines. A good understanding of the complexity of a specific data set arising from an application is a prerequisite for setting a proper expectation on classification performance, & to addressing questions like whether, or to what extent, patterns exist in a specific problem. It can also provide an important guide to the selection of classification techniques. However, there is a huge & diverse literature in this field, which can make it very difficult to keep up with latest advances. Classification is a central topic in several disciplines related to pattern recognition including machine learning, neural networks, & data mining, & it has numerous practical applications in image, video & speech signal processing, biometrics, digital libraries etc. Differences in approach and terminology often mean that different communities are unaware of concurrent developments under slightly difference names.This book is unique in its complete coverage & multidisciplinary approach. Classification & data characteristics are important issues that are all pervasive & impact upon real-world problems - researchers & practitioners alike will find this book an excellent reference source.
Synopsis
A unique book for researchers and practitioners across disciplines (including Computer Science, Engineering, Social Science and Finance) to learn about the current status, available techniques, and applications of data complexity in pattern recognition.
Synopsis
Machines capable of automatic pattern recognition have many fascinating uses in science and engineering as well as in our daily lives. Algorithms for supervised classification, where one infers a decision boundary from a set of training examples, are at the core of this capability. Tremendous progress has been made in refining such algorithms; yet, automatic learning in many simple tasks in daily life still appears to be far from reach. This book takes a close view of data complexity and its role in shaping the theories and techniques in different disciplines and asks: • What is missing from current classification techniques? • When the automatic classifiers are not perfect, is it a deficiency of the algorithms by design, or is it a difficulty intrinsic to the classification task? • How do we know whether we have exploited to the fullest extent the knowledge embedded in the training data? Data Complexity in Pattern Recognition is unique in its comprehensive coverage and multidisciplinary approach from various methodological and practical perspectives. Researchers and practitioners alike will find this book an insightful reference to learn about the current status of available techniques as well as application areas.
Synopsis
Automatic pattern recognition has uses in science and engineering, social sciences and finance. This book examines data complexity and its role in shaping theory and techniques across many disciplines, probing strengths and deficiencies of current classification techniques, and the algorithms that drive them. The book offers guidance on choosing pattern recognition classification techniques, and helps the reader set expectations for classification performance.
Table of Contents
Theory and Methodology.- Measures of Geometrical Complexity in Classification Problems.- Object Representation, Sample Size and Dataset Complexity.- Measures of Data and Classifier Complexity and the Training Sample Size.- Linear Separability in Descent Procedures for Linear Classifiers.- Data Complexity, Margin-based Learning and Popper's Philosophy of Inductive Learning.- Data Complexity and Evolutionary Learning.- Data Complexity and Domains of Competence of Classifiers.- Data Complexity Issues in Grammatical Inference.- Applications.- Simple Statistics for Complex Feature Spaces.- Polynomial Time for Complexity Graph Distance Computation for Web Content Mining.- Data Complexity in Clustering Analysis for Gene Microarray Expression Profiles.- Complexity of Magnetic Resonance Spectrum Classification.- Data Complexity in Tropical Cyclone Positioning and Classification.- Human-Computer Interaction for Complex Pattern Recognition Problems.- Complex Image Recognition and Web Security.