Analysis of Diabetes Disease Risk Prediction and Diabetes Medication Pattern Based on Data Mining

Abstract

Diabetes mellitus is the second most common disease after cardiovascular diseases and malignant tumors. With the continuous acceleration of people's living standards and life rhythm, the number of diabetic patients is rapidly increasing and showing a trend of youthfulness. A recent study found that 114 million adults in China have diabetes and have a high prevalence rate, but the awareness rate, treatment rate, and compliance rate are low. If diabetes is not treated and controlled in time, various complications will occur, such as cardiovascular, cerebrovascular, and diabetic foot, which will not only have a great impact on the survival of the patient, but also cause a lot of pressure on the family and society. Therefore, prevention and control of diabetes is an important strategy to save medical resources and reduce medical costs. In this paper, we mainly read a lot of literature and accumulate some important theoretical knowledge to clarify the basic principles and methods of data mining and refer to the research results of other scholars to select a new combined algorithm model combining K-means algorithm and logistic regression algorithm to construct a prediction model of diabetes and explore the law of medication for diabetic patients based on this analysis.