Rule Extraction for Diagnosis of Diabetes Mellitus Used for Enhancing Regular Covering Technique

Diabetes is a complex disease then increasing in prevalence around the world. Type 2 diabetes mellitus (T2DM accounts for about 80-90% of all diagnosed adult cases of diabetes. Most present diagnostic methods for T2DM are black-box models, which are unable to provide the reasons underlying diagnosis to physicians therefore, algorithms that can provide further insight are needed. Rule extraction can provide such explanations. The extracted rules may not be only highly accurate, but also simple and easy to understand. Therefore in the study, The rule extraction algorithm is proposed for Enhancing regular covering technique(ERCT) to achieve highly accurate, concise, and interpretable classification rules for the pima Indian diabetes(PID) dataset, which comprises 768 samples with two classes(diabetes or non-diabetes) and eight continuous attributes. The result of the proposed algorithm is an average accuracy of 85.55% after 10 runs of 10-fold cross validation. Regular covering technique achieved substantially better accuracy and provided a considerably fewer average number of rules and antecedents. These results suggest that proposed algorithm, and are therefore more suitable for medical decision making including the diagnosis of all type of diabetes mellitus.