Multi-source food hazard event extraction for public health

Abstract

Early detection and dissemination of food hazard information plays a key role in food hazard prevention and control. However, people in general cannot place enough time and effort for discovering food related hazards. In this paper, we introduce a food hazard monitoring system that can automatically collect online news and social media, extract events related to food hazards, and then organize the data in a structured format for easy consumption. We define an information template for food hazard event based on data from the Korean Ministry of Food and Drug Safety (MFDS), and use the template to aggregate informative keywords from government reports, news and social media. We explain our work process, considerations, as well as our future work to improve our system.