Diagnosis of diabetes type-II using hybrid machine learning based ensemble model

Abstract

The work done in this paper exhibits an expert system based ensemble model in diagnosing type-II diabetes. Diabetes Mellitus is a disease with high mortality rate that affects more than 60% population. The mindset of this task is to analyze various machine learning techniques for binary classification concerning with illness i.e. to diagnose whether a subject is suffering from disease or not. There are in total fifteen classifiers considered and out of them five major techniques namely: ANN, SVM, KNN, Naive Bayes and Ensemble are used. For achieving the desired goals the tools that were employed namely matrix laboratory (MATLAB) and WEKA 3.6.13. In Ensemble method the predictive potentials of various individual classifiers are fused together. Using Ensemble method, it increases the performance by combining the classifying ability of individual classifiers and the chances of misclassifying a particular instance are reduced significantly, this provides a greater accuracy to the overall classification process. It is the enhancing technique that does the majority voting and gives us the percolated results. The medical database analysed in this study includes a rich database of about 400 people from across a wide geographical region and ten physiological attributes. Furthermore, this diagnostic tool is examined by verifying denary cross attestation; on top of that the outcome has been confronted along the truly existing real interpretation about the cases. A GUI based diagnostic tool founded upon ensemble classifier is developed in such a manner it would be able to predict whether a patient is enduring against the disease or not when it is fed with all the 10 attributes from user through a user friendly GUI (Graphical User Interface).The development of this diagnostic tool is done using MATLAB 2013a. Out of 10 parameters that the user needs to enter as input in GUI based diagnostic tool five are numeric and the rest are nominal values. The diagnostic tool in execution is demonstrated below in Fig. 3. The main objective of this manuscript is to propose an intelligent framework that will act as a useful aid for doctors for correct and timely biopsy can be done at early stage. The result indicated that ensemble technique assured an accuracy of 98.60% that clubs the predictive performance of multiple AI based algorithms and are superior in comparison with all other individual counterparts. The algorithms with better exactness than others are followed by Artificial neural network (ANN), Naïve Bayes, Support Vector Machine (SVM), K-Nearest Neighbor (K-NN).