Soft computing based system modeling & control

Meena Tushir

ISBN 10: 365982366X / ISBN 13: 9783659823664

Verlag: LAP LAMBERT Academic Publishing Jan 2016, 2016

Beschreibung

This item is printed on demand - Print on Demand Titel. Neuware -Several clustering algorithms have been explained in the book including kernel based clustering algorithm. A kernel based clustering incorporates a kernel metric in place of the Euclidean distance used in the objective function. The kernel induced metric maps the data points to a high dimensional feature space, in which the data is more clearly separable, thereby increasing the accuracy of the proposed clustering technique. A fuzzy controller can also be designed using the clustering based approach. Clustering-based rule extraction methods help avoid combinatorial explosion of rules with increasing dimension of the input space. Also, because clustering step provides good initial rule parameter values, the subsequent rule parameter optimization process usually converges quickly and to a good solution.Books on Demand GmbH, Überseering 33, 22297 Hamburg 96 pp. Englisch.

Bestandsnummer des Verkäufers 9783659823664

Bibliografische Details

• Titel: Soft computing based system modeling & control
• Verlag: LAP LAMBERT Academic Publishing Jan 2016
• Erscheinungsjahr: 2016
• Sprache: Englisch
• ISBN-10: 365982366X
• ISBN-13: 9783659823664
• Einband: Taschenbuch
• Zustand: Neu

Über diesen Titel

Inhaltsangabe

Several clustering algorithms have been explained in the book including kernel based clustering algorithm. A kernel based clustering incorporates a kernel metric in place of the Euclidean distance used in the objective function. The kernel induced metric maps the data points to a high dimensional feature space, in which the data is more clearly separable, thereby increasing the accuracy of the proposed clustering technique. A fuzzy controller can also be designed using the clustering based approach. Clustering-based rule extraction methods help avoid combinatorial explosion of rules with increasing dimension of the input space. Also, because clustering step provides good initial rule parameter values, the subsequent rule parameter optimization process usually converges quickly and to a good solution.