Performance Evaluation of Classification Methods for Predicting Heart Disease

EasyChair Preprint 5955

Abstract

In this modern era industries are generating large amount of data. The most important asset of any kind of organization is data. Digitization reduces human effort and makes data easily accessible. With the increased access to the data, all industries are using data for making decisions. Health care industries are one among the top in generating data. To mine the complex data advance algorithms and techniques are needed. The data extraction techniques are used to convert these raw facts as meaningful information. One of the popular data extraction techniques is data mining and machine learning. With the patient data Health care industries are now focusing on optimizing the efficiency and quality of the treatment using various data analytical tools. Data mining and Machine learning has been used by many industries, however they are the proven methodology in health care. Non communicable disease such a heart disease, diabetics and cancer are major reason for the death around the world. Heart disease is one among the top reason for death. In this research paper we have implemented popular data mining algorithms viz., Support vector machine and decision tree with the relevant heart disease data set using Python. The performance of the algorithms is evaluated using various evaluation metrics.

Keyphrases: Data Mining, Decision Tree, Heart Disease, Support Vector Machine, health care, machine learning

@booklet{EasyChair:5955,
  author    = {S. Poonguzhali and P. Sujatha and P. Sripriya and V. Deepa and K. Mahalakshmi},
  title     = {Performance Evaluation of Classification Methods for Predicting Heart Disease},
  howpublished = {EasyChair Preprint 5955},
  year      = {EasyChair, 2021}}