An Ensemble Classifier Characterized by Genetic Algorithm with Decision Tree for the Prophecy of Heart Disease

Abstract

The prediction of heart disease is critically significant for diagnosis of diseases and treatment. The data mining techniques that can be applied in medicine, and in particular some machine learning techniques including the mechanisms that make them better suited for the analysis of medical databases. Extensive amounts of data gathered in medical databases require specialized tools for storing and accessing data, for data analysis, and for effective use of data. In particular, the increase in data volume causes great difficulties in extracting useful information and also consumes time for decision support. Intuitively, this large amount of stored data contains valuable hidden information, which could be used to improve the decision making process of an organization. For prediction of heart disease, many researchers have used some machine learning algorithms like Bayesian Classification, Neural Networks, Support Vector Machines, and K-nearest neighbor algorithms. We propose a hybrid technique of ensemble classifier to provide a better solution for the classification problem. The output from this hybrid scheme gives the optimized feature. This output is then given as the input to the decision tree classifier for predicting the occurrence and possibly obtaining the type of heart disease. Here, the features are initialized through the decision tree and fitness is evaluated via genetic algorithm.