“A Novel Approach Prediction of Heart Disease using Ordinary decision Tree, J48, Random Forest, Random Tree and Rep Tree in Weka Tool”

COMPUTER ENGINEERING

The Healthcare industry collects large amounts of Healthcare data, but unfortunately not all the data are mined which is required for discovering hidden patterns and effective decision making. Discovery of hidden patterns and relationships often goes unexploited. Advanced data mining modeling techniques can help remedy this situation. Cardiovascular disease is the leading cause of death in many countries. Health problems are enormous in this recent situation because of the prediction and the classification of health problems in different situations. The data mining area included the prediction and identification of abnormality and its risk rate in these domains. Identifying the major risk factors of Heart Disease categorizing the risk factors in an order which causes damages to the heart such as high blood cholesterol, diabetes, smoking, poor diet, obesity, hypertension and stress. Some of the key and most common techniques for data mining are association rules, classification, clustering, prediction, and sequential models. For a wide range of applications, data mining techniques are used. In our research we explore with different classification trees like ordinary decision tree (ODT), Hoeffding tree, J48 (C4.5), Random Forest (RF), Random tree (RT) and REP tree for rule induction in order to identify high quality data set. To execute of classification decision trees, we will use WEKA tool on the dataset. Finally results of all trees are compared using the performance measures like accuracy, correct and incorrect instances and execution times. Based on validation results on the data set, Random forest (RF) and Random tree (RT) are highest correct instances and in case of root mean square error (RMSE), Random tree (RT) is on top with lowest error rate as compare to others decision tree algorithms. .

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.