On the sensitivity of the probability of error rule for feature selection

The low sensitivity of the probability of error rule (Pe rule) for feature selection is demonstrated and discussed. It is shown that under certain conditions features with significantly different discrimination power are considered as equivalent by the Pe rule. The main reason for this phenomenon li...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on pattern analysis and machine intelligence. - 1979. - 2(1980), 1 vom: 01. Jan., Seite 57-61
1. Verfasser: Ben-Bassat, M (VerfasserIn)
Format: Aufsatz
Sprache:English
Veröffentlicht: 1980
Zugriff auf das übergeordnete Werk:IEEE transactions on pattern analysis and machine intelligence
Schlagworte:Journal Article
LEADER 01000naa a22002652 4500
001 NLM216985773
003 DE-627
005 20231224033023.0
007 tu
008 231224s1980 xx ||||| 00| ||eng c
028 5 2 |a pubmed24n0723.xml 
035 |a (DE-627)NLM216985773 
035 |a (NLM)22499623 
040 |a DE-627  |b ger  |c DE-627  |e rakwb 
041 |a eng 
100 1 |a Ben-Bassat, M  |e verfasserin  |4 aut 
245 1 0 |a On the sensitivity of the probability of error rule for feature selection 
264 1 |c 1980 
336 |a Text  |b txt  |2 rdacontent 
337 |a ohne Hilfsmittel zu benutzen  |b n  |2 rdamedia 
338 |a Band  |b nc  |2 rdacarrier 
500 |a Date Completed 02.10.2012 
500 |a Date Revised 12.11.2019 
500 |a published: Print 
500 |a Citation Status PubMed-not-MEDLINE 
520 |a The low sensitivity of the probability of error rule (Pe rule) for feature selection is demonstrated and discussed. It is shown that under certain conditions features with significantly different discrimination power are considered as equivalent by the Pe rule. The main reason for this phenomenon lies in the fact that, directly, the Pe rule depends only on the most probable class and that, under the stated condition, the prior most probable class remains the posterior most probable class regardless of the result for the observed feature. A rule for breaking ties is suggested to refine the feature ordering induced by the Pe rule. By this tie-breaking rule, when two features have the same value for the expected probability of error, the feature with the higher variance for the probability of error is preferred 
650 4 |a Journal Article 
773 0 8 |i Enthalten in  |t IEEE transactions on pattern analysis and machine intelligence  |d 1979  |g 2(1980), 1 vom: 01. Jan., Seite 57-61  |w (DE-627)NLM098212257  |x 1939-3539  |7 nnns 
773 1 8 |g volume:2  |g year:1980  |g number:1  |g day:01  |g month:01  |g pages:57-61 
912 |a GBV_USEFLAG_A 
912 |a SYSFLAG_A 
912 |a GBV_NLM 
912 |a GBV_ILN_350 
951 |a AR 
952 |d 2  |j 1980  |e 1  |b 01  |c 01  |h 57-61