Feature Selection of Gene expression data for Cancer Classification using SCF with SVM

Feature Selection is the technique used to select the features from microarray dataset. The selected feature must provide high accuracy to the intended classifier. An ordinary microarray dataset will have the characteristics such as high dimensionality limited sample and a large amount of noisy data. This basic features of microarray dataset will reduce the classification accuracy and elevate the run time of the proposed algorithm. To overcome this problem, dimensionality reduction techniques are deployed on the proposed dataset. Different dimensionality reduction techniques are available and they can be mainly categorised as feature selection and feature extraction. This research work focuses mainly on the filter based feature selection method. The proposed filter based combination method for performing dimensionality reduction is named as Feature selection of Gene expression data for Cancer Classification by using Score based Criteria Fusion (SCF) with SVM. The primary aim of the proposed research work is to minimize the classification time and make significant progression in the accuracy of algorithm.