An Outcome Based Analysis on Heart Disease Prediction using Machine Learning Algorithms and Data Mining Approaches

• Published in: 2022 IEEE World AI IoT Congress (AIIoT) pp. 01 - 07
• Main Authors: Deb, Aushtmi, Akter Koli, Mst. Sadia, Akter, Sheikh Beauty, Chowdhury, Adil Ahmed
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.06.2022
• Subjects: Classification algorithms; Data mining; Decision Tree; Feature extraction; Heart; Heart Disease; K-Nearest Neighbor; Logistic Regres-sion; Machine learning algorithms; Naive Bayes; Ran-dom Forest; Random forests; Support Vector Machine; Support vector machine classification
• DOI: 10.1109/AIIoT54504.2022.9817194

We have analyzed the World Health Organization report and found that over 17 million people have died of heart attacks in the last few decades. The diagnosis of heart-related disease and ensuring proper treatment is becoming immensely difficult due to the massive rise in population. However, recent advancement in technology has accelerated the health sector significantly. This paper aims at data mining techniques and analyse the various machine learning algorithms like Naive Bayes, Random Forest Classification, Decision tree, K-Nearest Neighbor, Logistic Regression, and Support Vector Machine by using a suitable dataset for heart disease prediction. Our findings suggest that Random forest provides the best possible prediction compared to others. One more conclusion from the research is that the decision tree has also shown better accuracy with the help of the Bagging ensemble method and k-fold cross-validation.