作者: Jin-Hyuk Hong , Sung-Bae Cho
DOI: 10.1016/J.PATREC.2005.07.009
关键词:
摘要: With increasing interest in bioinformatics, sophisticated tools are required to efficiently analyze gene information. The classification of expression profiles is crucial those fields. Since the features data obtained by microarray technology come be over thousands, it essential extract useful information selecting proper features. without any feature selection might redundant so that this can deteriorate performance classification. conventional method with genetic algorithm has difficulty for huge-scale selection. In paper, we modify representation chromosome suitable and adopt speciation enhance obtaining diverse solutions. Experimental results DNA from cancer patients show selected genes proposed