A Feature Importance Measure Based on Similarities of Feature Values

被引：0

作者：

Li, Yan ^{[1
]}

Zhang, Xiuli ^{[1
]}

Wang, Huachao ^{[1
]}

机构：

[1] Hebei Univ, Key Lab Machine Learning & Computat Intelligence, Coll Math & Comp Sci, Baoding City 071002, Hebei Province, Peoples R China

来源：

2011 AASRI CONFERENCE ON APPLIED INFORMATION TECHNOLOGY (AASRI-AIT 2011), VOL 1 | 2011年

关键词：

Nominal feature; Similarity; Importance Measure; Relative positive field; PSO; FEATURE-SELECTION;

D O I：

暂无

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

The preliminary work [1-2] of this paper originally proposed an importance measure of features based on the similarity between feature values, in which the nominal features are discussed. This paper further study the feasibility of the proposed feature importance measure and compare this measure with those based on entropy and based on Rough Set Theory.

引用

页码：362 / 365

页数：4

共 15 条

[1] Bagherjeiran A, 2008, STUD COMPUT INTELL, V73, P91
[2] Cui Y., 2007, J HENAN POLYTECHNIC, V26, P463
[3] FEATURE SELECTION WITH A LINEAR DEPENDENCE MEASURE
DAS, SK
[J]. IEEE TRANSACTIONS ON COMPUTERS, 1971, C 20 (09) : 1106 - &
[4] Hervás R, 2006, LECT NOTES ARTIF INT, V4106, P151
[5] Kennedy J., 1995, 1995 IEEE International Conference on Neural Networks Proceedings (Cat. No.95CH35828), P1942, DOI 10.1109/ICNN.1995.488968
[6] Li Y, 2005, LECT NOTES COMPUT SC, V3776, P780
[7] Li Y., 2010, ICMLC, V4, P1868
[8] [刘文军 LIU Wen-jun], 2009, [数学的实践与认识, Mathematics in Practice and Theory], V39, P148
[9] Unsupervised feature selection using feature similarity
Mitra, P
Murthy, CA
Pal, SK
[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2002, 24 (03) : 301 - 312
[10] Nguyen N. B., 2000, LECT NOTES COMPUTER, V1910, P13

← 1 2 →