This book discusses several approaches to obtaining knowledge concerning the performance of machine learning and data mining algorithms. It shows how this knowledge can be reused to select, combine, compose and adapt both algorithms and models.
From the reviews:
"There are many techniques available for machine learning from data ... . the problem is: given a set of data, which of the learning systems should one use? The goal of this book is to initiate a study of this problem. ... The mixture of detailed description and overview is well managed. The reader is able to see how the authors’ ideas and work fit into a larger framework. Graduate students looking for thesis topics should read this book." (J. P. E. Hodgson, ACM Computing Reviews, May, 2009)