Heart Disease Prediction System Using Naïve Bayes

Abstract

A large amount of data is generated in medical organisations (hospitals, medical centres) but this data is not properly used. There is a wealth of hidden information present in the datasets. This unused data can be converted into useful data. For this purpose, we can use different data mining techniques. This paper presents a classifier approach for the detection of heart disease and shows how Naive Bayes can be used for classification purposes. In our system, we will categorise medical data into five categories namely no, low, average, high and very high. Also, if an unknown sample comes then the system will predict the class label of that sample. Hence two basic functions namely classification (training) and prediction (testing) will be performed. The accuracy of the system depends on the algorithm and database used.