Introduction to Clustering Large and High-Dimensional Data

摘要

Jacob Kogan presents a variety of rigorous approaches to tackling one of the most interesting problems in almost all disciplines - that of discovering naturally arising structures in data. The book is divided into 10 main sections, with the eleventh providing solutions to selected problems. Its structure clearly aims at providing a theoretical understanding of data clustering algorithms - a worthy goal; bearing in mind that with so many data clustering algorithms in use, the focus should be on developing unified and optimal approaches, and this is what the book seeks to achieve. In particular, the six pages of Section 9 epitomise the ultimate goals of any data clustering specialist - labelling and assessing the clusters.