Enhanced Multicategory Classification Using Fast SVM Learning (FSVML) for Microarray Gene Expression Cancer Diagnosis

The main objective of this paper is to develop a fast and efficient classification method called the FSVML approach for a multicategory cancer diagnosis problem based on microarray data. ANOVA ranking has been used for ranking the input gene data. From the ANOVA ranking technique, top ranked genes are selected. This ranking technique improves the overall performance of the classification approach. Moreover, MLM learning is integrated with the SVM classifier. Thus, FSVML integrated system provides better results when compared with the other two approaches. Three datasets are used namely Lymphoma, Leukemia and SRBCT.