Diagnosis of Diabetes Using Support Vector Machine and Ensemble Learning Approach

Chitra Arjun, Mr.Anto S

In recent years, support vector machines (SVMs) have shown good performance in a number of application areas. The existing system is concentrated on the discovery of risk of having pre-diabetes or undiagnosed diabetes and to facilitate people decide whether they should see a physician for further evaluation.  It is also focused on both the noninvasive and metabolic factors, which should require blood sampling and laboratory measurements, such as high density lipoprotein (HDL), and cholesterol (CHOL). However the existing system ahs issue with prediction results by using c4.5, naïve bayes tree and neural network algorithms. To avoid the above mentioned issue we go for proposed system. In proposed scenario, we introduced an efficient algorithm named as Support Vector Machine (SVM) which is utilized to screen diabetes, and an ensemble learning module is added. It turns the “black box” of SVM decisions into comprehensible and transparent rules, and it is also useful for solving difference problem. The proposed system is used to develop an ensemble system for diabetes diagnosis. Specifically, the rules are extracted from the SVM algorithm and it is applied to provide comprehensibility and transparent representation. These rule sets can be regarded as a second opinion for diagnosis and a tool to screen the individuals with undiagnosed diabetes by lay users. From the experimental result, we can conclude that the proposed system is better than the existing scenario in terms of reduction of the incidence of diabetes and its complications.