This book provides an introduction to support vector machines (svm) and other kernel-based learning methods. it is written by nello cristianini and john shawe-taylor, and published by cambridge university press. the book begins with an overview of the learning methodology, including supervised learning, learning and generalisation, and the attractions and drawbacks of learning. it then moves on to discuss linear learning machines, including linear classification and regression, and the dual representation of linear machines. the next chapter explores kernel-induced feature spaces, including the implicit mapping into feature space, making kernels, and the relationship between kernels and gaussian processes. the book then delves into generalisation theory, including probably approximately correct learning, vapnik-chervonenkis theory, and margin-based bounds on generalisation. it also covers optimisation theory, including problem formulation, lagrangian theory, and duality. the chapter on support vector machines discusses support vector classification and regression, as well as their implementation. the book also includes a chapter on implementation techniques, including gradient ascent, chunking and decomposition, and sequential minimal optimisation (smo). the final chapter discusses applications of svm in text categorisation, image recognition, hand-written digit recognition, and bioinformatics. the book also includes pseudocode for the smo algorithm, background mathematics, and a list of references and an index. the book is intended for readers with a basic understanding of machine learning and mathematics, and it provides a comprehensive overview of svm and other kernel-based learning methods.This book provides an introduction to support vector machines (svm) and other kernel-based learning methods. it is written by nello cristianini and john shawe-taylor, and published by cambridge university press. the book begins with an overview of the learning methodology, including supervised learning, learning and generalisation, and the attractions and drawbacks of learning. it then moves on to discuss linear learning machines, including linear classification and regression, and the dual representation of linear machines. the next chapter explores kernel-induced feature spaces, including the implicit mapping into feature space, making kernels, and the relationship between kernels and gaussian processes. the book then delves into generalisation theory, including probably approximately correct learning, vapnik-chervonenkis theory, and margin-based bounds on generalisation. it also covers optimisation theory, including problem formulation, lagrangian theory, and duality. the chapter on support vector machines discusses support vector classification and regression, as well as their implementation. the book also includes a chapter on implementation techniques, including gradient ascent, chunking and decomposition, and sequential minimal optimisation (smo). the final chapter discusses applications of svm in text categorisation, image recognition, hand-written digit recognition, and bioinformatics. the book also includes pseudocode for the smo algorithm, background mathematics, and a list of references and an index. the book is intended for readers with a basic understanding of machine learning and mathematics, and it provides a comprehensive overview of svm and other kernel-based learning methods.