This document discusses using genetic algorithms for feature selection and support vector machines for classification to improve prediction of heart disease. It first reviews literature on using various machine learning techniques for heart disease prediction, including support vector machines, neural networks, random forests, naive Bayes classifiers and decision trees. The methodology section then outlines the steps taken, which include collecting a dataset on heart disease from a public repository, preprocessing the data using genetic algorithms to select important features, and classifying the reduced data using naive Bayes, random forest and support vector vector machine classifiers to predict heart disease. The goal is to select key features and improve prediction accuracy.