Risk Level Prediction for Heart Disease using Decision Tree Induction

Abstract

Heart Disease was the major cause of causalities in most of the countries. According to the medical records, heart disease kills one person in very sort time. Classification and prediction are the forms of data analysis that can be used to extract models for important classes or to predict future data trends. In this paper, decision tree induction algorithm is used to classify the risk level for heart disease. Decision tree is a flow-chart-like tree structure, where each internal node denotes a test on an attributes, each branch represents an outcome of the test, and the leaf nodes represent classes or class distributions. This system generates the understandable rules for user and estimates the accuracy for classifier. Depending on the attribute values of the data set, this system can classify the risk level of heart disease whether it is in serious or normal conditions for patients. Thus, the user can test his or her medical check concerned with their heart. Moreover, the system can provide the classifier accuracy by using Holdout Method.