A Hybrid Data Mining Approach For Type-2 Diabetes Prediction And Classification

Diabetes is one of the typical metabolic diseases that can lead to debilitating effects on the human body. Controlling the disease requires changing eating patterns, and in the absence of timely treatment can lead to very serious conditions in the patient. According to statistics, around 77 million people diagnosed with diabetes in India, meanwhile, China with 196 million It has the highest statistics in the world. Therefore, the diagnosis and treatment of disease using advanced methods can be very useful in reducing mortality and timely treatment. In the present research study, a novel scheme based on the data mining methods to predicate type 2 diabetes going to propose, and try to improve the accurateness of diabetes prediction techniques by making a hybrid model and use of different datasets to achieve a better result. The hybrid model is based on the Artificial Neural network (ANN) and Support Vector Machine (SVM) and it compared with the K-means and logistic regression algorithm. The analysis of data is done with the Weka tool and pre-processing shows improvements with consideration of already available research studies result. The performance of the present study from a different aspect shows new advances in the diagnosis of diabetes.