An improved Computer Aided Clinical Decision Making System for classifying the tumor has been developed and presented in this paper. The texture and shape features extracted from preprocessed mammograms have been utilized to obtain the optimal multiple  feature sets using multiobjective genetic algorithm. The Multilayer Back Propagation Neural Network (MBPN), Self Organising Map(SOM) with major voting method have been used to classify the tumor as benign or malignant. The multiple features with optimal feature selection is found to have the diagnostic accuracy 99.5%. The performance of the proposed clinical decision support system has been estimated and found that this system will provide valuable information to the physicians in clinical pathology.