M.A. Thesis:

The use Ensemble methods for Prediction of Heart disease

Cardiovascular disease is one of the most important causes of mortality in the world and is one of the major health problems in developed and developing countries. Data mining as a technique for identifying and diagnosing diseases and categorizing patients in disease management and finding patterns for quicker diagnosis of patients and preventing complications in them can be of great help. The purpose of this study is to predict heart disease using combination methods. These methods are used to combine classroom techniques. Three classification algorithms called neural network, Decision tree and Naive Bayes were evaluated and the results of all three methods were evaluated using cross-validation method. It was then combined with two combined methods including the majority vote and the nearest neighbor in the IBK, and the final result of the superiority of the IBK method was to the majority vote in the combination of the three methods mentioned. In the present study, data from heart disease Data Sets from the site UCI was used.

Keywords: Heart disease prediction - Neural network - Majority vote - Nearest neighbor - IBK - Combined method