Systemy uczące się with delivery to Ireland

This book is devoted to methods of building machine learning programs: programs capable of improving their performance on the basis of past experience. Programs of this kind acquire knowledge automatically and then use it to carry out the tasks they are given.

The author introduces the reader to the learning process of systems from an algorithmic point of view. He explains what inductive learning involves and presents the two main approaches to the subject: decision tree induction and decision rule induction. He discusses probabilistic learning methods as well as conceptual clustering methods, and deals with the problems of transforming a set of attributes. He presents selected algorithms for learning function approximation, and describes the discovery of dependencies in data, the learning of automata, and the problem of reinforcement learning.

The book's substantive value is enhanced by numerous exercises, divided into traditional, laboratory and project exercises.