Optimization of Logistic Regression Model by Feature Engineering using Fuzzy Logic

In modern era, the Machine learning especially Supervised Machine Learning is iridescent field where many researchers are shaping their research work. Innovations and new ideas in field are growing fast. Still in various fields like medical, financial and other, we need to enhance the predictions of the Logistic Regression classification models. In this research paper, we focused on the optimization of the supervised Logistic Regression model which can predict more accurately when implemented on clinical dataset. Our purpose is to introduce and implement the fuzzy logic for Features engineering technique to optimize the logistic regression model before training of the Logistic regression, which can predict more accurately in the medical data sets. In this paper, we use Coronary Heart Disease (CHD) for implementing the features engineering using fuzzy logic. Final result showed that Optimized Logistic Regression Model prediction of Ten Years Heart risk is improved with accuracy of. 863 when classical Logistic Regression classification model has been compared with it.