Redundant Feature Selection Based on Hybrid GA and BPSO

Redundant Feature selection is an important topic in the field of bioinformatics. This paper proposes a novel algorithm on Redundant Feature Selection Based on Hybrid GA and BPSO(RFS-GSO), which tries to find a compact feature subset with great predictive ability. Compared with the previous works, RFS-GSO measures the redundancy of feature set by the maximum feature inter-correlation, which is more reasonable than those by the averaged inter-correlation. The outstanding performance of RFS-GSO has been examined by the experiments on several real world microarray data sets.