A Comparative Study on Application of Class-Imbalance Learning for Severity Prediction of Adverse Events Following Immunization
This work addresses a domain-specific problem for medical practitioners by providing a prediction system to improve severity assessment of adverse vaccine reactions, though it is incremental as it builds on existing class-imbalance learning methods.
The study tackled predicting hospitalization for adverse events following immunization in children using imbalanced data, and found that an improved RUSBoost algorithm achieved the highest Area Under the ROC Curve among tested methods.
In collaboration with the Liaoning CDC, China, we propose a prediction system to predict the subsequent hospitalization of children with adverse reactions based on data on adverse events following immunization. We extracted multiple features from the data, and selected "hospitalization or not" as the target for classification. Since the data are imbalanced, we used various class-imbalance learning methods for training and improved the RUSBoost algorithm. Experimental results show that the improved RUSBoost has the highest Area Under the ROC Curve on the target among these algorithms. Additionally, we compared these class-imbalance learning methods with some common machine learning algorithms. We combined the improved RUSBoost with dynamic web resource development techniques to build an evaluation system with information entry and vaccination response prediction capabilities for relevant medical practitioners.