Metalearning: Applications to Data Mining - Brossura

Brazdil, Pavel

Brossura

ISBN 10: 3642092314 ISBN 13: 9783642092312

Casa editrice: Springer, 2010

Riassunto

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.

Su questo libro

Recensione

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)

Contenuti

Metalearning: Concepts and Systems.- Metalearning for Algorithm Recommendation: an Introduction.- Development of Metalearning Systems for Algorithm Recommendation.- Extending Metalearning to Data Mining and KDD.- Extending Metalearning to Data Mining and KDD.- Bias Management in Time-Changing Data Streams.- Transfer of Metaknowledge Across Tasks.- Composition of Complex Systems: Role of Domain-Specific Metaknowledge.

Dettagli dell?edizione

• Editore: Springer
• Data di pubblicazione: 2010
• ISBN 10: 3642092314
• ISBN 13: 9783642092312
• Rilegatura: Copertina flessibile
• Lingua: Inglese
• Numero di pagine: 192
• Contatto del produttore: non disponibile