Abstract

Abstract: The technology allows users to utilize algorithms to forecast the likelihood of diabetes mellitus in their bodies and forecast whether or not they will get cancer. The several categorization models are employed in this system, such as Decision Tree, artificial neural networks, logistical regression, association rules and the Naive Bay. The Random Forest approach is then used to discover the exactness of each model in the project. The employed dataset is a Pima Indians Diabetes Data Set with the information of patients, some of whom acquire diabetes, which makes the project a mobile application designed to forecast whether or not the class of a person is at risk for diabetes and cancer. We are studying four common illness risk prediction classifiers. These algorithms consist of Decision Tree, Artificial Neural Network, Regression Logistic and Bay Naive. After that, strategies for bagging and boosting using these algorithms are integrated to increase the solidity of eachmodel. At the end, the algorithm of Random Forest is used. The aim of this study is to estimate the risk of cancer and diabetes without blood testing or hospitalization for everybody. The study is also aimed at encouraging and promoting healthy human health.Keywords: Artificial neural network(ANN);Super Vector Machine(SVM);Extreme Learning Machine(ELM);Decision Tree (DT) ;Random Forest(RF);Convolution Neural Network(CNN);Confusion Matrix(CM)