Optimized Heart Disease Prediction through Fuzzy Rough Set based Missing Data Imputation Techniques

Abstract

This abstract describes an effort to include fuzzy-rough set-based missing data imputation approaches into heart disease prediction models to improve their accuracy. Missing values are a major difficulty and may jeopardize the accuracy of predictive studies in medical datasets, such as those kept in repositories like the University of California, Irvine (UCI). This paper suggests using fuzzy-rough set-based imputation techniques, which are well-known for their ability to handle the ambiguity and uncertainty included in medical data, to address this problem. Fuzzy-rough sets and their latest expansions are used to introduce the Cardiovascular Disease Multiple Imputation Technique (CVDMIT), a unique technique. By way of extensive testing using a Random Forest classifier, CVDMIT is thoroughly analyzed and compared with well-known methods like as fuzzy roughest, fuzzy C means, and expectation maximization. The results show that using CVDMIT in conjunction with Random Forest classification results in a significant improvement in accuracy, with a precision rate of 94%. With increased accuracy and dependability, this study advances the field by highlighting the potential of fuzzy-rough set-based missing data imputation approaches in optimizing heart disease prediction.