A research framework for passive surveillance for food safety from social media: Identification and evaluation of customer reviews for regulatory use and case study of 30 restaurants

Abstract

The primary objective of this paper is to develop a framework for continuous monitoring of the safety of food business operators without overburdening established regulatory systems through social media for food safety. A phase-wise methodology was adopted, wherein Phase 1 was dedicated to identifying available literature on Adverse Drugs Reactions (ADR) reporting using Social Media data. Phase 2 used the data from google maps review of the restaurants to replicate a similar methodology for Food Safety Surveillance. We identified 5 themes for a complete Surveillance framework, theme 1 involves data collection from social media, theme 2 involves pre-processing of data for analysis, theme 3 involves data annotations, theme 4 involves Identifying the relationship between regulatory violation and event, and theme 5 involves evaluation of the model. We were able to demonstrate the ADR reporting methodology could be adopted till theme 3, whereas theme 4 requires the development of an algorithm to assess the causality of an event with the Food Safety Code. According to our research, it is possible to develop a passive surveillance system for food safety that adheres to the principle of ADR reporting; however, the main obstacle is the absence of a causality assessment algorithm that can link an event to the food safety code and help regulators take immediate action.