Integrating Multiple Data Sources and Learning Models to Predict Infectious Diseases in China
Published in AMIA 2019 Summit, 2018
Goal: infectious disease(flu, HFRS, mumps etc.) morbidity rate prediction. More specific: compared to traditional infeactious disease prediction which mainly focus on historical morbidity incidences, our research uses multimodal deep learning, combining info from morbidity history, weather, air quality and search engine/ social network trend. And the result(avg MAPE ~12%) greatly outperforms traditional ML method(ARIMA, xgboost etc). Role: three-month research, working as the major contributor.