http://www.reviews.com/review/review_review.cfm?review_id=130914

Computing Reviews
	
	
Today's Issue 	Hot Topics 	Search 	Browse 	Recommended 	My Account 	Log In
Review 	Help 			
	
		
	
	
Search 		
	
	
	
Introduction to machine learning (Adaptive Computation and Machine Learning Series)
Alpaydin E., The MIT Press, 2004. Type: Book
Date Reviewed: Mar 4 2005 	
	
	

This monograph provides students, researchers, and application developers with the knowledge and tools required to get the most out of the theory of mathematical statistics, and related data modeling methods for solving a large variety of machine learning problems. Since the main aim of machine learning is to obtain computer programs able to use samples of data, and to include knowledge derived from past experience in solving problems of great difficulty, work in machine learning converges from many sources, including mathematical statistics, neural networks, brain models, adaptive control theory, psychological models, artificial intelligence, and evolutionary models.

Alpaydin supplies a comprehensive and sound introduction to some of the most significant sub-fields of machine learning. The concepts and methods are presented in a very clear and accessible way, and the illustrative examples provided by the author are welcome, and prove extremely helpful to readers. A list of bibliographic references, citing the most representative titles for each topic, is given at the end of each chapter.

The content is structured into 16 chapters. The fundamentals of supervised learning, Bayesian decision theory, and different parametric methods are presented in the opening chapters. Aspects related to the dimensionality reduction problem, such as principal component analysis, factor analysis, and multidimensional scaling, are briefly exposed in the sixth chapter of the book. The next four chapters are focused on nonparametric density estimation methods, clustering, decision trees, linear discrimination, and several related topics. Neural computing offers a viable alternative to the Von Neumann computational paradigm in solving some particular classes of problems; these provide a suitable framework for obtaining more powerful computing systems.

Artificial neural network structure multilayer perceptrons (MLPs), and several variants of the gradient descent training algorithm, are presented in the eleventh chapter of the monograph. The final four chapters are devoted to some of the most specific topics in machine learning, namely, hidden Markov models (HMM), techniques for improving learning performance by combining more learners, and reinforcement learning.

In my opinion, the content of the book is outstanding, with regard to the clarity of the discourse and the variety of well-selected examples. The enlightening comments provided by the author at the ends of the chapters, and the suggestions for further reading, are also important features of the book.

Although primarily a textbook for teaching undergraduate and postgraduate courses in machine learning, this book is also of interest, and helpful, to practitioners and researchers.
Reviewer:  L. State 	Review #: CR130914
		
  	Reviewer Selected
	 
  	
Learning (I.2.6 ) 	
	 
  	
Deduction And Theorem Proving (I.2.3 ) 	
	 
  	
Heuristic Methods (I.2.8 ... ) 	
	 
  	
Probability And Statistics (G.3 ) 	
	 
  	
Problem Solving, Control Methods, And Search (I.2.8 ) 	
	 
  	
Statistical Computing (G.3 ... ) 	
	 
  	more 	 
	
		
	
Would you recommend this review? 	
	yes 	
	no 	
	
		
	
Other reviews under "Learning": 	Date
Learning to decode cognitive states from brain images
Mitchell T., Hutchinson R., Niculescu R., Pereira F., Wang X., Just M., Newman S.  Machine Learning 57(1-2): 145-175, 2004. Type: Article 	Mar 11 2005
Information theory, inference and learning algorithms
MacKay D.,  Cambridge University Press, New York, NY, 2002. 550 pp. Type: Book, Reviews: (2 of 2) 	Jan 17 2005
A hybrid language model based on a combination of N-grams and stochastic context-free grammars
Linares D., Benedí J., Sánchez J.  ACM Transactions on Asian Language Information Processing 3(2): 113-127, 2004. Type: Article 	Jan 14 2005
more...
	
		
	E-Mail This Printer-Friendly 	
		
	
	
[Reviewer's area] 	[Masthead] 	[Subscribe to Reviews.com] 	[Press Releases] 	[Tips] 	[Help] 	Contact Us
	
	
Reproduction in whole or in part without permission is prohibited.   Copyright 2005 Reviews.com™
Terms of Use | Privacy Policy