-
Views
-
Cite
Cite
Gilles Celeux, Data Clustering: Theory Algorithms and Applications by Gan, G., Chaoqun, M. A., and Wu, J., Biometrics, Volume 64, Issue 2, June 2008, Pages 656–657, https://doi.org/10.1111/j.1541-0420.2008.01026_7.x
Close - Share Icon Share
Extract
This book gives a tour of cluster analysis methods and algorithms. The authors present a considerable number of techniques and programs for clustering using a direct, fairly terse approach to the subject. Many of the described techniques have been proposed recently in conference proceedings. The authors provide an algorithmic point of view, and many algorithms are presented in pseudo code.
The first part of the book (six chapters) describes basic concepts (type of data, coding, data transformation and visualization, distances, and so forth). This part is rather complete, but provides little advice for choosing appropriate ways to describe data.
The second part, which is the core of the book (11 chapters), is devoted to clustering algorithms. Classical clustering methods are presented, along with more recent approaches such as subspace clustering for high-dimensional databases. The authors put some emphasis on fuzzy clustering and optimization techniques. The organization of this important part of the book is good. I was just a little bit surprised to see the chapter on fuzzy clustering presented before the chapter on center-based clustering algorithms. (Consequently, the fuzzy k-means algorithm is presented before the k-means algorithm.)
