Heart Disease Prediction Using Association Rule Mining

Description

Heart Disease is one of the crucial diseases in humans. The risk of heart disease can lead to death, therefore, timely detection and prevention of heart disease is an essential task for medical science. However, with the increasing growth of technology and machine learning, it becomes possible to analyze human activity and daily routine to identify the risk of heart disease risk. The machine learning has the ability to explore the data and identify the fruitful insights from the data. In this presented work, a heart disease prediction system has been introduced by using machine learning techniques. In this context, an improved apriori algorithm has been implemented by modifying the traditional apriori algorithm. First, the input data has been preprocessed and the outliers have been identified. Next the outliers have been eliminated and then the data has been encoded. The encoding is performed to convert the number values into a symbols to represent the itemsets. After encoding of data, the transaction sets have been prepared and then a modified Apriori algorithm has been implemented to recover the decision rules. The modified algorithm aims to reduce the number of rules generated by the apriori algorithm, and results in the most potential rules for predicting heart disease. The implementation has been done using JAVA technology. Additionally, to store the measured performance the MySQL server has been used. The experiments have been done on the Heart Disease dataset obtained from Kaggle. Additionally, to simulate the effectiveness of the proposed modified apriori algorithm a comparison between traditional apriori and the proposed apriori algorithm has been performed in terms of accuracy, error rate, training time, and memory usage. Based on the evaluated parameters of proposed and traditional system, it was found the proposed technique is effective and accurate as compared to the traditional apriori algorithm-based heart disease prediction system.

The running example of the system is given as: