Diabetes prediction model construction method and system based on machine learning

Abstract

The invention relates to a diabetes prediction model construction method and system based on machine learning, and the method comprises the following steps: S1, obtaining the glycometabolism data of a sample population, and constructing a first sample data set; S2, preprocessing the obtained initial sample data set to obtain a second sample set, and dividing the second sample set into a training set and a verification set according to a preset proportion; S3, selecting a plurality of single-classification machine learning models to construct a primary model, and taking the training set as input for training to obtain a plurality of single disease prediction models; S4, performing diversity analysis on the single disease prediction model, and selecting two models with large difference and highest precision for fusion to obtain a fusion model; S5, constructing a comprehensive prediction model based on the single disease prediction model and the combined model. According to the method, the model precision and generalization ability are effectively improved, the probability that the to-be-detected person suffers from the diseases related to the abnormal glucose metabolism can be rapidly predicted, and nervous medical resources are saved.

Description

technical field [0001] The invention relates to the field of big data analysis, in particular to a method and system for constructing a diabetes prediction model based on machine learning. Background technique [0002] Type 2 diabetes is the main type of diabetes, and the danger lies in its complications. Diabetes and its complications have become the main cause of death and disability among the population, and seriously threaten people's health. It should be noted that type 2 diabetes is a preventable and controllable chronic disease, and its development often takes a period of time. Due to the complexity and multifactorial nature of diabetes and prediabetes, the prevention of diabetes and prediabetes must refer to multiple risk factors. However, it can be difficult to assess a person's risk of abnormal glucose metabolism when there are many predictors present at the same time. A diabetes risk prediction model that integrates multiple related factors will help promote he...

AI Technical Summary

Problems solved by technology

The existing traditional systems and models cannot accurately analyze the corresponding data and realize data prediction, and the model accuracy and generalization ability are weak, so they cannot be applied to existing prediction equipment

Images