2017/05/22 by Julio Albinati, Albinati, Julio, Wagner Meira +7
Medicine · Physics and Astronomy · Social Sciences · #Complex Network Analysis Techniques #Data-Driven Disease Surveillance #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Mosquito-borne diseases and control #Social and Information Networks (cs.SI)
paper · pdf · doi:10.48550/arxiv.1705.07879
openalex publication_date 2017/05/22 · openalex created_date 2022/10/02 · openalex updated_date 2026/07/28
Epidemiological early warning systems for dengue fever rely on up-to-date\nepidemiological data to forecast future incidence. However, epidemiological\ndata typically requires time to be available, due to the application of\ntime-consuming laboratorial tests. This implies that epidemiological models\nneed to issue predictions with larger antecedence, making their task even more\ndifficult. On the other hand, online platforms, such as Twitter or Google,\nallow us to obtain samples of users' interaction in near real-time and can be\nused as sensors to monitor current incidence. In this work, we propose a\nframework to exploit online data sources to mitigate the lack of up-to-date\nepidemiological data by obtaining estimates of current incidence, which are\nthen explored by traditional epidemiological models. We show that the proposed\nframework obtains more accurate predictions than alternative approaches, with\nstatistically better results for delays greater or equal to 4 weeks.\n