Analysis of Machine Learning Models for Heart Disease Prediction using Different Algorithms
Main Article Content
Abstract
This research work seeks to explain the development of existing research on utilizing computational intelligence techniques in heart diseases diagnosis. This disorder extremely malignant ailment, over 1.7 billion demise all over the world. Diagnosis and treatment of heart diseases at early stage is the only solution otherwise it leads to fatality rate. With pace of time, new technologies emerging such as AI &ML, IoT and due to these advancement in science especially in healthcare, various types of severe disease can be diagnosed at early stage. The main objective of this work is to design machine learning models to predict heart disease with better accuracy. In our implemented work five different supervise ML (Machine Learning) algorithms are instigated which are Logistic Regression, KNN, SVM, Decision Tree and Bagging Classifier. Out of listed algorithm, SVM perform better and give the accuracy 93.40% and KNN gives the least accuracy 71.42%. Accuracy in machine learning models should not be so high otherwise it will be fall under over fitting. Machine learning models having accuracy more than 90 % is measured upright. One important thing is that accuracy should not be so high otherwise it may be possible that designed model is overfit for a specific dataset. Besides accuracy in this research article two parameters also calculated which are precision and recall from confusion matrix. Support vector machine algorithm gives precision value 88 and recall value 91.67.