A Hypertension Risk Prediction Model Based on Improve Random Forest

In order to reduce the serious consequences of chronic diseases, this paper proposes a hypertension risk prediction model based on improved random forest, which provides an effective technical means for early warning of hypertension. The original data set with unbalanced samples is processed by the synthetic minority oversampling technique (SMOTE) to form a balanced data set. Then improve the random forest algorithm based on similarity optimization and deep optimization, and finally establish a prediction model. It is compared with the four machine learning algorithms of linear regression (LR), artificial neural network (ANN), support vector machine (SVM) and CatBoost. ROC curve and AUC are used as the evaluation indicators of the model. The experimental results show that the prediction accuracy of the model based on the improved random forest algorithm is higher, with an AUC value of 0.8697, which is better than the other four algorithms. The improved random forest algorithm has certain feasibility in hypertension risk prediction. This method has a better effect in predicting the risk of hypertension, which is better than other traditional methods, can provide more accurate judgments, and provide better results for early warning and prevention of hypertension.