The objective of the work aims to formulate a regression model to predict the occurrence of the heart disease using minimum number of parameters. The problem of heart disease is chosen owing to the increasing risk of heart disease in India. The Data is collected from a hospital (Cleveland Dataset) which consists of 22 attributes and a class label as retrieved from UCI Repository. Technique of Step wise regression is used to formulate this model and this enables us to identify the model of the highest accuracy of about 89.72% that contains the attributes which have the highest effect on the class label (outcome variable). The technique of supervised learning is used to train and obtain the mathematical model. The observed result is an expression function value of the selected attributes that corresponds to the determination of heart disease with fewer set of attribute with its parametric values. Thereby, the number of tests that corresponds to the disease can be reduced which then reduces the expenses towards the disease and its co-morbidities.
CITATION STYLE
Harsheni, S. K., Souganthika, S., Gokul Karthik, K., Sheik Abdullah, A., & Selvakumar, S. (2020). Analysis of the Risk Factors of Heart Disease Using Step-Wise Regression with Statistical Evaluation. In Lecture Notes on Data Engineering and Communications Technologies (Vol. 35, pp. 712–718). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-32150-5_70
Mendeley helps you to discover research relevant for your work.