Diagnosis of Diabetes Type-II Using Ensemble Model with Adaptive Neuro Fuzzy Inference System (ANFIS), and Improved Support Vector Machine Algorithm (ISVM)
Use of blood sugar (glucose) by body is affected by a group is disease termed as Diabetes mellitus. Prediction of Diagnosis of type-II diabetes is a complicated task. Different machine learning approaches are used to solve this issue with illness. Prediction classifies patient with disease. Artificial Neural Network (ANN), Naive Bayes (NBs), K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) is used as ensemble classification techniques.
It is very important to enhance the classifier performance in detection. Improved Support Vector Machine Algorithm (ISVM) as well as Adaptive Neuro Fuzzy Inference System (ANFIS) is introduced in this paper to enhance the accuracy. Majority voting function is used to combine these two classifiers (ANFIS and SVM) and it is termed as ANFIS+ISVM.
Pima Indians Diabetes Database (PIDD) dataset is used to implement the proposed classifier ANFIS+ISVM. There are two parts in the proposed prediction model. They are testing and training. MATLAB tool is used for experimenting ANFIS+ISVM classifier. Recall, F-Measure, Accuracy and Precision are used for evaluating the performance.