Machine Learning Classification Model to Identify Heart Disease

Abstract

Cardiac disease has been considered a complicated condition that has an enormous impact on a major number of individuals throughout the world. In medicine, especially in cardiology, a timely and precise diagnosis of the condition of the heart to look out for the problems mandatory. Therefore, in the paper, we came up with a method for diagnosing heartrelated sickness which uses regressive machine learning models and algorithms, so that it ensures that the model works efficiently and tries to be as accurate as possible. For reducing non-related and use- less characteristics, the model relies on classification methods that comprise. As we need to surpass the feature selection issue, the latest and quick has been used. The features selection methodology helps us to improve the accuracy and decrease the implementation time used by the classification technique of the model. In ac- quisition, the employed to find best practices in the evaluation of the model and hyperparameter tweaking. To ensure the classifier’s execution accuracy value measures have been utilized. The effectiveness of the classifiers has been checked by characteristics that have been selected through features extraction techniques. The practical findings suggest that using a classifier support vector machine to cre- ate a supreme intellectual model to detect cardio-related illness, an affordable fea- ture selection technique (FCMIM) has been recommended. In comparison to earlier offered approaches, the recommended diagnostic system (FCMIM - SVM) obtained high accuracy. Furthermore, the suggested approach may be simply used in the medical field to detect cardiac problems.