Matrix Methods in Data Mining and Pattern Recognition
Buy a book... In Association with Amazon.co.uk
Author(s): L. Elden
Publisher: SIAM
ISBN: 9780898716269
Format: softback
224pp
Price: $69.00
Review Date: 25 March 2008
Review: The book has been developed for an undergraduate course on numerical algorithms for data mining and IT at Linkoping University and is aimed at undergraduates who have taken a scientific course or recent graduates in data mining and pattern recognition who want an introduction to linear algebra techniques. There are powerful techniques for this but because the book is application oriented to book does not give a comprehensive treatment of the mathematical and numerical aspects. The book is in three parts. The first covers an introduction, linear algebra concepts and matrix decomposition. From this MATLAB should be able to be used. In part two linear algebra is applied to data mining and problems are chosen which are well suited to the techniques. A short introduction to eignevalue and singular value algorithms are given in part three. Rather than give a book of recipes the author has intended to give the reader a set of tools to be tried as they are, but may likely need to be modified for a particular application. A collection of exercises and computer assignments are given at the book’s webpage.