Pattern Classification by Richard Duda, Peter Hart, and David Stork renders the principal algorithms of pattern classification with the help of low range MATLAB objects. The building of objects is illustrated in a systematic manner using visualizations and data building algorithms with the help of solved programs. The design of algorithm has been explained in a series of chapters with the initial chapters on Bayesian concepts like Bayesian Decision Theory and Parameter Estimation along with the concept of Maximum-Likelihood. Different statistical methods and uni-dimensional discriminant mappings have been discussed in the book. The two different methods such as stochastic and non-metric methods have been explained along with learning techniques- unsupervised and algorithm-independent machine learning techniques, including clustering. An appendix and an index section mark the concluding pages of the book.
· Bayesian Decision Theory
· Maximum-Likelihood and Bayesian Parameter Estimation
· Nonparametric Techniques
· Linear Discriminant Functions
· Multilayer Neural Networks
· Stochastic Methods
· Nonmetric Methods
· Algorithm-Independent Machine Learning
· Unsupervised Learning and Clustering
14 Mar, 2015
23 Sep, 2014
Flipkart wanted me to read Book in 30 days.
22 Sep, 2014
kind of ok
25 Mar, 2014