A technical analysis on data mining classification algorithms for CVD prediction

Cardio Vascular Disease (CVD) is seemingly increasing for various reasons including life style modification and heredity. Multiple techniques and research are going for, how to predict and detect CVD’s. Mostly, Blood Pressure, Cholesterol and Pulse rate are some of important parameters involved in these checkouts. However, viewing the age factor of human’s these factors vary in tandem. Research follow up on CVD prediction indicates two main aspects i) Accuracy of CVD prediction is seemingly high if more attributes are taken into consideration and ii) Computer assisted methods are very much required for both observation and exact prediction. Present study on CVD detection dictates the need for computational decision support system for enhancing medical experts to wisely decide on clinical process and also to reduce risk over traditional support system. The present study aims to compare and analyze the different Data Mining classification algorithms such as Decision Tree, J48, Naïve Bayes, Bayes Net, Simple Cart, Artificial Neural Network, Genetic Algorithms, Nearest Neighbour, Associative classification, Generalized Linear Models, Support Vector Machine, C5, K-Means, MPSOK-means, Logistic Regression, ID3, and K-Nearest Neighbour which have been used in the prediction of CVD’s.