A novel BPSO approach for gene selection and classification of microarray data

• Published in: 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence) Vol. 10; pp. 2147 - 2152
• Main Authors: Yang, Cheng-San, Chuang, Li-Yeh, Li, Jung-Chike, Yang, Cheng-Hong
• Format: Conference Proceeding Journal Article
• Language: English
• Published: IEEE 01.06.2008
• Subjects: Artificial neural networks; Joints
• ISBN: 1424418208; 9781424418206; 9781424432196; 1424432197
• ISSN: 2161-4393; 1522-4899; 2161-4407
• DOI: 10.1109/IJCNN.2008.4634093

Selecting relevant genes from microarray data poses a huge challenge due to the high-dimensionality of the features, multi-class categories and a relatively small sample size. The main task of the classification process is to decrease the microarray data dimensionality. In order to analyze microarray data, an optimal subset of features (genes) which adequately represents the original set of features has to be found. In this study, we used a novel binary particle swarm optimization (NBPSO) algorithm to perform microarray data selection and classification. The K-nearest neighbor (K-NN) method with leave-one-out cross-validation (LOOCV) served as a classifier. The experimental results showed that the proposed method not only effectively reduced the number of gene expression levels, but also achieved lower classification error rates.