Diabetes Prediction Algorithm Using Recursive Ridge Regression L2

At present, the prevalence of diabetes is increasing because human body cannot metabolize glucose level. Accurate prediction patients an important research area. Many researchers have proposed techniques to predict this disease through data mining and machine learning methods. In prediction, feature selection a key concept in preprocessing. Thus, features that are relevant used for prediction. This condition improves accuracy. Selecting right whole set complicated process, many concentrating on it produce predictive model with high work, wrapper-based method called recursive elimination combined ridge regression (L2) form hybrid L2 regulated algorithm overcoming overfitting problem set. Overfitting major selection, where new unfit training small. Ridge mainly overcome problem. The selected by using method, random forest classifier classify basis features. work uses Pima Indians Diabetes set, evaluated results compared existing algorithms prove accuracy algorithm. predicting 100%, its area under curve 97%. outperforms algorithms.