Multilevel Risk Prediction of Cardiovascular Disease based on Adaboost+RF Ensemble Learning

2019 The 5th International Conference on Electrical Engineering, Control and Robotics

无线电电子学;计算机科学

Li, Runchuan^1^2 ; Shen, Shengya^3 ; Chen, Gang^1^2^4 ; Xie, Tiantian^1 ; Ji, Shasha^1 ; Zhou, Bing^1^2 ; Wang, Zongmin^1^2

Industrial Technology Research Institute, Zhengzhou University, Zhengzhou Henan

450000, China^1

Cooperative Innovation Center of Internet Healthcare, Zhengzhou University, Zhengzhou Henan

450000, China^2

School of Foreign Languages, Zhengzhou University, Zhengzhou Henan

450000, China^3

School of Distance Learning, Zhengzhou University, Zhengzhou Henan

450000, China^4

关键词: Cardio-vascular disease;    Contribution degree;    Ensemble learning;    Information gain ratio;    Multiple levels;    Prediction model;    Risk predictions;    Unbalanced datasets;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/533/1/012050/pdf
DOI  :  10.1088/1757-899X/533/1/012050

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Background: In the field of diagnostic CVD, the predecessors used a large amount of data with no missing two-category data, and obtained good results. However, in the process of electronic input of historical data, a large number of data attribute values are missing, and there are multiple levels of disease risk. Goal: On the data set of imbalance and a large number of missing values, this paper focuses on the five levels of cardiovascular disease. Methods: A new prediction model of Adaboost+RF is constructed by using the information gain ratio to analyze the feature contribution degree of the data set. The performance of this model is evaluated with Precision, Recall, F-measure and ROC Area values. Results: The results show that the four key indicators of the Adaboost+RF model on five-categories unbalanced datasets in Precision, Recall, F1 and AUC values, which are 40.9%, 49.3%, 41.4% and 71.6%. Conclusion: The experiment results demonstrate that the four key indicators of the Adaboost+RF model on five-category unbalanced missing datasets are better than other machine learning algorithms.