An Improved Method for Diabetes Prediction through the Application of Neural Network

A common disease that is affecting the whole world is diabetes, and is also known as a silent killer. Diabetes affects the nervous system, retina of the eyes, kidney, heart and affects the entire system of the body. Diabetes boosts the other diseases also whenever patients suffer from them otherwise it remains in sleep mode. The foremost reason for boosting diabetes is the lifestyle of today’s generation. In this paper, we have tried to predict diabetes at the primary stage by making use of a neural network i.e., multilayer perceptron. We have made use of the PIMA Indians diabetes dataset in our article and the experiments were performed using the Weka tool. After applying the proposed algorithm, the experimental result shows an increase in accuracy by 3% which is far better for predicting diabetes.

Keywords: Accuracy, Diabetes, Neural Network, Performance.