Building a Robust Heart Diseases Diagnose Intelligent Model based on RST using LEM2 and MODLEM2

Abstract:

Rough set theory (RST) is the effective and powerful approach to rules induction from data/decision tables. It based on discovering knowledge in form of rules. These rules used to predict and classify the new cases from included decision cases or states. Knowledge discovery from the medical database is very useful to improve prediction and diagnosing of diseases risk levels. In this paper RST based algorithms Modified Learning from Examples Module, version 2 (MODLEM2) are used to used extract a valuable set of rules that detect the heart risk pathologies state. MODLEM2 with ENTROPY and LAPLACE measures are used generate an extended minimal of rules without the need for prior preprocessing. MODLEM2 with its measure allow discretization performed and rule induction at the same time. The results of Entropy-MODLEM2 and Laplace MODLEM2 compared with the LEM2 algorithm which module of Learning from Examples based on Rough Sets (LERS) that generate minimal rules. The presented results of the intelligent techniques ensure that DOMLEM2 with its two measures are better than LEM2 in number or rules generated and classification accuracy.