A Prediction Model With Machine Learning In Cardiovascular Disease Risk

According to a report from the World Health Organization,cardiovascular disease is a high-risk chronic disease.At present,the prevalence of cardiovascular diseases in our country is not optimistic with death accounting for a large proportion and the economic burden of treatment heavy.With the development of information technology,the research on cardiovascular disease prevention has gradually become a cross-cutting issue in various fields such as medical treatment,computer,and statistics.Meanwhile,the electronic health record system of our country has gradually developed.How to use the Chinese text data recorded by the system to establish a prediction model of cardiovascular disease has become a problem for more and more researchers.Based on the above background,this paper selects coronary heart disease as a representative cardiovascular disease and introduces the GBDT feature extraction method to replace the general data preprocessing process.Meanwhile,this paper uses the GBDT feature extraction method to test on the balanced datasets and the unbalanced datasets of coronary heart disease,verifying that the GBDT feature extraction method is essective in coronary heart disease risk prediction.Then,this paper uses algorithms related to natural language processing to establish a corpus based on Chinese cardiovascular disease text data,and builds a text data structured model,and tests it on disserent classifiers to obtain accuracy.Finally,it proves The reliability of the model.The GBDT feature extraction proposed in this paper can extract hidden information in the data preprocessing step,and it can generally improve the model essect of cardiovascular disease prediction models.It has high applicability.Meanwhile,the text classification model proposed in this article has a good classification essect on small corpus and small sample sets.It can help the electronic information health system to further mine and use the recorded Chinese text data,to help potential patients with cardiovascular disease successfully to achieve the purpose of prevention.