Does Machine Learning Make More Reliable Diabetes Prediction Model Using Big Data?: Comparison of Gradient Boosting Decision Tree and Logistic Regression

Abstract We sought to verify the reliability of machine learning (ML) in developing diabetes prediction models by utilizing big data. To this end, we compared the reliability of gradient boosting decision tree (GBDT) and logistic regression (LR) models using data obtained from the Kokuho-database of the Osaka prefecture, Japan. To develop the models, we focused on 16 predictors from health checkup data from April 2013 to December 2014. A total of 277,651 eligible participants were studied. The prediction models were developed using a light gradient boosting machine (LightGBM), which is an effective GBDT implementation algorithm, and LR. Their reliabilities were measured based on expected calibration error (ECE), negative log-likelihood (Logloss), and reliability diagram. Similarly, their classification accuracies were measured in the area under the curve (AUC). We further analyzed their reliabilities while changing the sample size for training. Among the 277,651 participants, 15,900 (7978 males and 7922 females) were newly diagnosed with diabetes within three years. LightGBM (LR) achieved an ECE of 0.0017±0.00033 (0.0048±0.00058), a Logloss of 0.167±0.00062 (0.172±0.00090), and an AUC of 0.845±0.0025 (0.826±0.0035). From sample size analysis, the reliability of LightGBM became higher than that of LR when the sample size increased more than 10^4. Thus, we confirmed that GBDT provides a more reliable model than LR in the development of diabetes prediction models using big data. ML could potentially produce a highly reliable diabetes prediction model, a helpful tool for improving lifestyle and preventing diabetes.