Heart Disease Diagnosis Using Machine Learning Techniques In Python: A Comparative Study of Classification Algorithms For Predictive Modeling

Abstract

Coronary illness is the essential driver of death these days. Medicines for coronary illness patients have been progressed, for instance with machine-to-machine (M2M) innovation to empower far off persistent observing. To utilize M2M to take care of distant coronary illness patient, his/her ailment ought to be estimated intermittently at home. Consequently, it is challenging to perform complex tests which need doctors to help. In the meantime, coronary illness can be anticipated by investigating a portion of patient's wellbeing boundaries. With assistance of information mining strategies, heart sickness expectation can be gotten to the next level. There are a few calculations that have been utilized for this reason like Guileless Bayes, Choice Tree, and k- Closest Neighbor (KNN). This study plans to utilize information mining procedures in coronary illness expectation, with improving boundaries to be utilized, so they can be utilized in M2M far off quiet checking reason. KNN is utilized with boundary weighting strategy to further develop precision. Just 8 boundaries are utilized (out of 13 boundaries suggested), since they are straightforward and moment boundaries that can be estimated at home. That's what the outcome shows the precision of these 8 boundaries utilizing KNN calculation are great enough, contrasting with 13 boundaries with KNN, or much other calculations like Guileless Bayes and Choice Tree.

Keywords: Heart Disease Prediction; k-Nearest Neighbor; Data Mining; Machine to Machine

JEL Classification: O3