Machine Learning and Data Mining. Introduction to Principles and Algorithms
Recent advances in machine learning have had a strong impact on rapid developments in related areas like data analysis, knowledge discovery and computational learning theory. This (already well-established) research discipline has found many applications in expert and business systems, data mining, databases, game playing, text, speech and image processing, etc. However, due to the rather interdisciplinary character of this new field, many books on machine learning currently appearing are quite specialised to the considered discipline. In that sense, this book differs from several other excellent books on machine learning and data analysis, as its authors have succeeded in providing a detailed, yet comprehensive, overview of the field. The entire textbook consists of fourteen chapters and can be divided into five parts.
The introductory part of the book encompasses the first two chapters, which provide an historical overview of the field and outline philosophical issues related to machine and human learning, intelligence and consciousness. The following four chapters deal with the basic principles of machine learning, representation of knowledge, basic search algorithms used to search hypothesis space, and attribute quality measures used to guide the search. The next two chapters set practical guidelines for preparing, cleansing and transforming the processed data. Chapters 9 – 12 are devoted to a description of the respective learning algorithms: symbolic and statistical learning, artificial neural networks and cluster analysis. The last two chapters introduce formal approaches to machine learning: the problem of identification in the limit and computational learning theory.
The attached appendix reviews some theoretical concepts used in the book. Although oriented primarily towards advanced undergraduate and postgraduate students, each section of the textbook is relatively self-contained and only requires a background knowledge in calculus. The discussed topics are explained in an interesting way with a lot of illustrative figures and supporting graphs. Each chapter is accompanied with a summary of the concepts taught. For these reasons, this book represents both a valuable teaching resource for students and a good reference source of applicable ideas for a wide audience including researchers and application specialists interested in machine learning paradigms.