Improving supervised learning algorithms using min max scalar method to increase accuracy for diabetes prediction

Abstract

Data mining is most widely used techniques in medical area due to the huge volume of data available in medical sector, with help of huge volume of data, valuable and useful pattern of Information and knowledge extracted. The proposed system model is effectively classified diabetes disease dataset as well as heart disease dataset. The main goal of the study is to increase the accuracy of supervised learning algorithm like support vector machine, random forest and K nearest neighbour using normalisation techniques such as min Max scaler. I have used PIDD and Cleveland heart disease datasets for research work which is taken from UCI machine learning repository. The basic supervised learning algorithm is compared with basic algorithm having min Max scalar techniques on both the data set, so that is identify which is effective techniques for diabetes disease and heart disease prediction. newline Diabetes disease is not previously diagnosis and treated appropriately severe various side effect like kidney failure, cardiovascular disease, Visual impairment etc. To diagnosis the diabetes disease patient undergoes various costly and time-consuming test. For the past ten years heart disease has been the world s leading cause of death. so, it is very dangerous and risky and also people have to fight for their lives. Prevention may allow those who suffer from this disease to improve their lives. newline newlineThe objective of the study is to boost the accuracy of supervised learning algorithms such as K Nearest Neighbour, Support Vector Machine and Random Forest using the Min Max techniques on diabetes dataset. Data mining techniques are used for early prediction of any kind of disease. The proposed model to predict diabetes disease and heart disease data set with accuracy. For research work used TensorFlow machine learning open-source platform for classification purpose and train the machine learning model. Various tensorflow libraries are used. newline