Prediction of Score of Diabetes Progression Index Based on Logistic Regression Algorithm

In order to overcome the problems of traditional diabetes prediction, a disease index grading method based on logistic regression algorithm is proposed. According to the score of disease progression index one year later, the results are divided into two categories, namely, the index score is greater than or equal to 150 and less than 150, and the problem of which interval the target value will belong to is well applicable to the logistic regression model. Compared with using linear regression algorithm to predict the impact of a feature on the progression of diabetes, logistic regression algorithm is an effective attempt. The classification results based on the logistic regression model show that when the two selected features are S5 and S6, the classification accuracy can reach 75.7%.