A Filter Method for Feature Selection for SELDI-TOF Mass Spectrum

Trung-Nghia VU  Syng-Yup OHN 

Publication
IEICE TRANSACTIONS on Information and Systems  Vol.E92-D  No.2  pp.346-348
Publication Date: 2009/02/01
Online ISSN: 1745-1361
Print ISSN: 0916-8532
Type of Manuscript: LETTER
Category: Pattern Recognition
Keyword: 
feature selection,  filter,  high dimensions,  mass spectrum, 

Full Text: PDF(169.8KB)

Summary: 
We propose a new filter method for feature selection for SELDI-TOF mass spectrum datasets. In the method, a new relevance index was defined to represent the goodness of a feature by considering the distribution of samples based on the counts. The relevance index can be used to obtain the feature sets for classification. Our method can be applied to mass spectrum datasets with extremely high dimensions and process the clinical datasets with practical sizes in acceptable calculation time since it is based on simple counting of samples. The new method was applied to the three public mass spectrum datasets and showed better or comparable results than conventional filter methods.