A Machine learning-based prediction model for the heart diseases from chance factors through two-variable decision tree classifier

• Published in: Journal of intelligent & fuzzy systems Vol. 41; no. 6; pp. 5985 - 6002
• Main Authors: Wang, Y., Chu, Y.M., Khan, Y.A., Khan, Z.Y., Liu, Q., Malik, M.Y., Abbas, S.Z.
• Format: Journal Article
• Language: English
• Published: Amsterdam IOS Press BV 01.01.2021
• Subjects: Arteries; Data mining; Decision making; Decision trees; Developing countries; Heart; Heart diseases; High density; Illnesses; LDCs; Lipoproteins; Machine learning; Physical exercise; Prediction models; Residential density; Risk analysis; Tobacco; Weight reduction
• ISSN: 1064-1246; 1875-8967
• DOI: 10.3233/JIFS-202226

This paper addressed the prediction of heart sicknesses from hazard elements through a decision-making tree. We introduced the facts mining technique in public fitness to extract high-degree knowledge from raw data, which facilitates predicting heart diseases from risk factors and their prevention. The existing work intends to introduce a new risk element in heart diseases using novel data mining strategies. Latest actual international affected person’s information (e.g., smoking, area of residence, age, weight, blood stress, chest pain, low-density lipoproteins (LDL), high-density lipoproteins (HDL), block arteries became accrued by way of the use of questionnaire through direct interview technique from patients. Novel two-variable decision trees are constructed for coronary heart illness records primarily based on chance factors and ranking of risk elements. The results show a correct prediction of cardiovascular disease (CVD) from the risk factor if records on chance factors are available as direct results of this study, tobacco, loss of physical exercise, and weight-reduction plan play a vital role in predicting heart diseases, which is the most important reason for mortality in developing countries, especially in my country.