Authors : T. Keerthika and K. Premalatha
Abstract: Worldwide heart disease forecast has been a major research over the past decade since the major reason of death is due to heart disease. Numerous researchers combined fuzzy technique with some other technique for proficient classification purpose in order to predict the heart disease, since the fuzzy is proficient only if proper fuzzy rules are specified in the rule base. At this point, we have introduced a rough-fuzzy classifier that shared rough set theory with the fuzzy set. Generally, there are three main steps taken part in the rough-fuzzy classifier such as: rule generation using rough set theory, rule optimization using Artificial Bee Colony (ABC) and prediction using fuzzy classifier. At first, the discernability matrix is framing by the given database. Reduct and core analysis is used to recognize the relevant attributes from the discernability matrix after that fuzzy rules are generated from the rough set theory. After that the set of rule is optimized. Then, with the assist of fuzzy rules and membership functions, the fuzzy system is intended so that the prediction can be carried out within the fuzzy system intended. Finally, the experimentation is carried out by means of the Cleveland, Hungarian and Switzerland datasets. From the results, we ensure that the proposed rough-fuzzy classifier outperformed the previous approach by achieving the accuracy of 87% in Hungarian and 80% in Switzerland datasets.
T. Keerthika and K. Premalatha, 2016. Heart Disease Prediction System Using Optimal Rough-Fuzzy Classifier Based on ABC. Asian Journal of Information Technology, 15: 481-492.