Computational Intelligence Techniques for Diabetes Prediction

Abstract

This study is on “Computational Intelligence Techniques for Diabetes Prediction”. The main objective is to examine the performance of various Machine Learning algorithms in order to reduce the high cost of chronic disease diagnosis by prediction. A huge number of individuals in Bangladesh alone experience the ill effects of undiscovered or late-analyzed unending sicknesses, for example, Endless Kidney Ailment (CKD), Coronary illness, diabetes, Bosom Malignant growth and some more. Most of the time such types of disease diagnosis is very costly and complicated. Considering diabetes, early prediction of diabetes is an important issue in Health Care Services (HCS). So, there is a need of an application that can effectively diagnosis thousands of patient using medical specification. I examine different machine learning algorithms for predicting diabetes in real time by sketch and gather from concepts and tools in the field of machine learning. This work uses 4 classification techniques for diabetes prediction. Such as, Artificial Neural Network, Random Forest , Naive Bayes and Support Vector Machine. The performance of different classification technique was evaluated on different measurements technique. In addition, my present investigation for the most part centers around the utilization of restorative code information for infection forecast, and investigate distinctive courses for speaking to such information in my expectation calculations.