Reinforced Rough Set Theory Based on Modified MEPA in Classifying Cardiovascular Disease

Cardiovascular disease is one of chronic diseases that seriously threaten human health. It is important that use clinical data to classify the cardiovascular diseases for supporting medical diagnosis. In this paper, we propose a new approach to reinforce rough set theory for classifying problems, which contains 3 main procedures: (1) Convert discretized continuous data into unique corresponding linguistic value; (2) use the linguistic value to extract decision rules by IEM2 (learning from Examples Module, version 2) algorithm; and (3) utilize rule filter to improve rule quality. In empirical case study, we use a practical collected dataset, which contains 1068 patients´ clinical chemistry analysis data to illustrate the proposed approach. From the results, the accuracy, coverage, and number of rules show that the proposed approach outperforms the listing rough set approach.