Abstract

In this paper, we used data analytics to perform research on heart disease. Since more data is becoming available, the subject of heart disease prediction is still relatively new. Several researchers have examined it using different methodologies and techniques. We used data analytics to locate and predict the sufferers of illnesses. We initially carried out a pre-processing step in which we identified which attributes were the most important based on the correlation matrix using three data analytics methodologies (Decision tree, Random forest, SVM, and KNN) on data sets of different sizes. This enabled us to assess the precision and stability of each method. The datasets are categorised using medical parameters. To examine such factors, our system employs a data mining classification technique. The datasets are analysed in Python using machine learning methods, with the best model demonstrating the highest level of accuracy for heart disease.

Description