New Frontiers in Communication and Intelligent Systems
Classification and Prediction of Heart Disease: A Machine Learning Approach
Authors: Minal Chaudhari and Rohini Patil
Publishing Date: 12-11-2021
ISBN: 978-81-95502-00-4
Abstract
Heart condition refers to any condition affecting heart. There are many sorts, a number of which are preventable. Unlike disorder, which incorporates problems with the whole cardiovascular system, heart condition affects only the heart. During this research, the dataset is collected from UCI repository which brings together data from 4 other databases Cleveland, Hungary, Switzerland and long beach. Dataset contains 1025 patient with 14 attributes is employed with a set value. The results of proposed model are comparing with the previous model where the marginally changes in accuracy with feature selection attribute by using cfs evaluator using genetic search method. There the multi layer perceptron algorithm increases their accuracy 91.21% to 91.41%. In data preprocessing some outliers and extreme value within the dataset would be removed. The whole 869 instances were used for classification. An information gain evaluator method was performed on heart disease dataset that shows increasing the performance of classification accuracy. The naïve bayes, SVM, MLP, KNN, J48 gives the accuracy 81.93, 82.62, 91.59, 99.65, 98.84 respectively.
Keywords
Heart Disease Prediction, Machine Learning, Classification Algorithm, Feature Selection.
Cite as
Minal Chaudhari and Rohini Patil, "Classification and Prediction of Heart Disease: A Machine Learning Approach", In: Rahul Srivastava and Aditya Kr. Singh Pundir (eds), New Frontiers in Communication and Intelligent Systems, SCRS, India, 2021, pp. 151-156. https://doi.org/10.52458/978-81-95502-00-4-18