Classification Model for the Heart Disease Diagnosis

Abstract

Medical science industry has huge amount of data, but unfortunately most of this data is not mined to find out hidden information in data. Advanced data mining techniques can be used to discover hidden pattern in data. Models developed from these techniques will be useful for medical practitioners to take effective decision. In this research work, we have analyzed the performance of the classification rule algorithms namely PART based on K-Means Clustering algorithms. The k-means is the simplest, most commonly and good behavior clustering algorithm used in many applications. Firstly the preprocessed heart disease dataset is grouped using the K-means algorithm with the K =2 values on classes to cluster evaluation testing mode. After that data mining classification rule algorithms namely Projective Adaptive Resonance Theory are analyzed on clustered relevant dataset. In our studies 10-fold cross validation method was used to measure the unbiased estimate of the prediction model. Accuracy of K-Means Clustering, PART and PART based on K-Means Clustering are 81.08%, 79.05% and 84.12% respectively.