Number of the records: 1
Sequential Retreating Search Methods in Feature Selection
- 1.
SYSNO ASEP 0357268 Document Type V - Research Report R&D Document Type The record was not marked in the RIV Title Sequential Retreating Search Methods in Feature Selection Author(s) Somol, Petr (UTIA-B) RID
Pudil, Pavel (UTIA-B) RIDIssue data Praha: ÚTIA, 2010 Series Research Report Series number 2286 Number of pages 21 s. Language eng - English Country CZ - Czech Republic Keywords feature selection ; wrappers ; sequential search ; subset search ; method evaluation ; classifier performance ; pattern recognition Subject RIV BD - Theory of Information R&D Projects 1M0572 GA MŠMT - Ministry of Education, Youth and Sports (MEYS) IAA2075302 GA AV ČR - Academy of Sciences of the Czech Republic (AV ČR) CEZ AV0Z10750506 - UTIA-B (2005-2011) Annotation Inspired by Floating Search, our new pair of methods, the Sequential Forward Retreating Search (SFRS) and Sequential Backward Retreating Search (SBRS) is exceptionally suitable for Wrapper based feature selection. (Conversely, it cannot be used with monotonic criteria.) Unlike most of other known sub-optimal search methods, both the SFRS and SBRS are parameter-free deterministic sequential procedures that incorporate in the optimization process both the search for the best subset and the determination of the best subset size. The subset yielded by either of the two new methods is to be expected closer to optimum than the best of all subsets yielded in one run of the Floating Search. Retreating Search time complexity is to be expected slightly worse but in the same order of magnitude as that of the Floating Search. In addition to introducing the new methods we provide a testing framework to evaluate them with respect to other existing tools. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2011
Number of the records: 1