Developing a dengue forecast model using machine learning: A case study in China
importance 6/10
ai assisted
Mosquito-borne diseases and control infectious disease surveillance stage: research
Problem
Accurate and timely forecasts of dengue incidence in China were lacking despite expanded affected areas and increased incidence.
Why it matters
Dengue remains an important public health issue in China with expanding affected areas.
What the AI did
Multiple machine learning algorithms (SVR, step-down regression, GBM, negative binomial regression, LASSO) combined a Baidu dengue search index with climate factors to forecast weekly dengue cases in Guangdong.
Results
not stated in source
Impact
ML-based forecasts could enable earlier dengue warnings using search and climate data.
Record
| Venue | PLoS neglected tropical diseases |
|---|---|
| Published | 2017 |
| Work type | article |
| Authors | Pi Guo, Tao Liu, Qin Zhang, Li Wang, Jianpeng Xiao, Qingying Zhang, Ganfeng Luo, Zhihao Li et al. |
| Citations | 263 (OpenAlex, at import time) |
| References | 51 |
| Model or method | SVR, GBM, LASSO, negative binomial regression, step-down regression (compared) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1371/journal.pntd.0005973 |
| Open access | gold |
| Keywords | dengue, machine learning, forecasting, Baidu search index, epidemiology |
Importance rating
6
significant — thousands to millions affected
Abstract states dengue is an important public health issue in China with expanded affected areas and increased incidence; forecasting supports earlier warnings.