Adaptive mechanisms for classification problems with drifting data

This source preferred by Hamid Bouchachia

Authors: Sahel, Z., Bouchachia, A., Gabrys, B. and Rogers, P.

Editors: Apolloni, A., Howlett, R.J. and Jain, L.

http://www.springerlink.com/content/n0286t8411252664/?p=dabdbe745320484d9b56fb79525cb888&pi=17

Pages: 419-426

Publisher: Springer

Place of Publication: Berlin

ISBN: 978-3-540-74826-7

DOI: 10.1007/978-3-540-74827-4_53

Most work on supervised learning is undertaken on static problems. However, in many real world classification problems, the environment in which the classifiers operate is dynamic i.e. the descriptions of classes change with time. In this paper, the process of generating drifting data is introduced in order to assess two adaptive approaches that deal with dynamically changing data. These approaches are: retraining on evolving data set and incremental learning. The empirical evaluation has shown that both these approaches improve the performance compared to the non-adaptive mode though a number of outstanding research issues remain.

This data was imported from DBLP:

Authors: Sahel, Z., Bouchachia, A., Gabrys, B. and Rogers, P.

Editors: Apolloni, B., Howlett, R.J. and Jain, L.C.

http://www.informatik.uni-trier.de/~ley/db/conf/kes/kes2007-2.html

Journal: KES (2)

Volume: 4693

Pages: 419-426

Publisher: Springer

ISBN: 978-3-540-74826-7

DOI: 10.1007/978-3-540-74827-4_53

This data was imported from Scopus:

Authors: Sahel, Z., Bouchachia, A., Gabrys, B. and Rogers, P.

Journal: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Volume: 4693 LNAI

Issue: PART 2

Pages: 419-426

eISSN: 1611-3349

ISBN: 9783540748267

ISSN: 0302-9743

DOI: 10.1007/978-3-540-74827-4_53

Most work on supervised learning is undertaken on static problems. However, in many real world classification problems, the environment in which the classifiers operate is dynamic i.e. the descriptions of classes change with time. In this paper, the process of generating drifting data is introduced in order to assess two adaptive approaches that deal with dynamically changing data. These approaches are: retraining on evolving data set and incremental learning. The empirical evaluation has shown that both these approaches improve the performance compared to the non-adaptive mode though a number of outstanding research issues remain. © Springer-Verlag Berlin Heidelberg 2007.

This data was imported from Web of Science (Lite):

Authors: Sahel, Z., Bouchachia, A., Gabrys, B. and Rogers, P.

Journal: KNOWLEDGE-BASED INTELLIGENT INFORMATION AND ENGINEERING SYSTEMS: KES 2007 - WIRN 2007, PT II, PROCEEDINGS

Volume: 4693

Pages: 419-426

ISBN: 978-3-540-74826-7

ISSN: 0302-9743

The data on this page was last updated at 04:40 on November 19, 2017.