A Generalized Output-Coding Scheme with SVM for Multiclass Microarray Classification

Abstract

Multiclass cancer classification based on microarray data is described. A generalized output-coding scheme combined with binary classifiers is used. Different coding strategies, decoding functions and feature selection methods are combined and validated on two cancer datasets: GCM and ALL. The effects of these different methods and their combinations are then discussed. The highest testing accuracies achieved are 78% and 100% for the two datasets respectively. The results are considered to be very good when compared with the other researchers’ work.