Preliminary Cardiac Disease Risk Prediction Based on Medical and Behavioural Data Set Using Supervised Machine Learning Techniques

Objectives: The objective of the work is to detect the probable signs and symptoms which might further lead to detection of cardiac diseases using the learned system which is trained using data collected from previous patient. Methods/Statistical Analysis: The data set has been taken from reputed machine leaning data set repositories. The data has been cleaned, imputed and then the Outliers are removed before using them for training purpose. The Classification methods which are nothing but Supervised Learning of Machine Learning Technology is used to train the system. In this work Classification Tree, Naive Bayes, Random Forest and Support Vector Machine algorithms are used for training the Prediction System. The experiment has been conducted using the python based Data Mining tool called Orange and the scores have been evaluated. Findings: The comparison of precision of various supervised learning algorithms are analysed and has been found out that Classification Trees are efficient in Prediction. Application/Improvements: As future work the data of Hospitals and Health Research Institutes can be uploaded in Cloud and the data analysis can be done extensively to get an accurate Prediction which can be used across many hospitals and research institutes throughout the world.