Active Learning Using Fuzzy k-NN for Cancer Classification from Microarray Gene Expression Data

Abstract

AbstractClassification of cancer from microarray gene expression data is an important area of research in the field of bioinformatics and biomedical engineering as large amounts of microarray gene expression data are available but the cost of correctly labeling it prohibits its use. In such cases, active learning may be used. In this context, we propose active learning using fuzzy k-nearest neighbor (ALFKNN) for cancer classification. Active Learning technique is used to select most confusing or informative microarray gene expression patters from the unlabeled microarray genes, so that labeling on the confusing data maximizes the classification accuracy. The selected most confusing/informative genes are manually labeled by the experts. The proposed method is evaluated with a number of microarray gene expression cancer datasets. Experimental results suggest that in comparison with traditional supervised k-nearest neighbor (k-NN) and fuzzy k-nearest neighbor (fuzzy k-NN) methods, proposed active learning method (ALFKNN) provides more accurate result for cancer prediction from microarray gene expression data.KeywordsActive learningCancer classificationMicroarray gene expression data(Fuzzy) k-nearest neighbor