One of the most prevailing and serious disease affecting human's health is Heart Disease (HD). Early diagnosis may allow for heart disease prevention or reduction, which could lower the rate of death.Machine Learning techniques have produced a variety of solutions for heart disease prediction and is capable of predicting illness at early stage . This study propose a model that includes many machine learning (ML) techniques to obtain accurate heart disease (HD) 2pDreedpiacrttimonesn.tDoaftMacaothlleemctaitoincsa,nIsdlapmrei-apCrooclelesgsienUgnairveeurssietdy,toPecsrheaawteaar,cKcuhryabteerdPaatakhfoturntkhhewtraa,iPnainkgismtaondel. Supervised Machine learning classifiers like support ve(fcotonrt 1m0atcihmineen(eSwVMro)m, adnecision tree (DT), logistic regression (LR), K Nearest Neighbor (KNN) and Naïve Bayes (NB) are used for predicting heart disease. (Moonestlrineleevspaanctifnegat)ures are selected by using Relief and LASSO feature selection techniques. Various evaluating methods like, sensitivity, accuracy, specificity, MCC,confusion matrix and precision are used for the performance evaluation of model. This study did comparative analysis using supervised machine learning and feature selection techniques. Decision tree gives highest accuracy of 85.21% with all features. On the other hand, with feature selection techniques SVM has an excellent performance. Future strategy is to use Deep learning algorithms and other feature selection techniques.