A PSO-SVM FEATURE SELECTION METHOD FOR IMPROVING THE PREDICTION ACCURACY OF HEART DISEASE

The primary concern of feature selection techniques in data mining is to extract task relevant data from voluminous datasets. Searching meaningful data in high dimensional datasets is trivial but results in enhanced performances over time and accuracy. Evolutionary algorithms play an important role in selecting the best features ever, than the traditional ones, as its primary goal centre around yielding optimized solutions. Thus, in this paper particle swarm optimization based feature selection algorithm is implemented on heart disease dataset to improve disease prediction.