Booster in High Dimensional Data Classification using CNN and Decision Tree Algorithm
Classification problems in high dimensional data with small number of observations are becoming more common especially in microarray data. The performance in terms of accuracy is essential while handling sensitive data particularly in medical field. For this the stability of the selected features must be evaluated. Therefore, this paper proposes a new evaluation measure that incorporates the stability of the selected feature subsets and accuracy of the prediction. Booster in feature selection algorithm helps to achieve the same. The proposed work resolves both structured and unstructured data using convolution neural network based multimodal disease prediction and decision tree algorithm respectively. The algorithm is tested on heart disease dataset retrieved from UCI repository and the analysis shows the improved prediction accuracy.