A Novel Parallelized LSTM For Detecting Internet Food Safety

Abstract

Food safety is a major problem concerning people's livelihood. With the advent of the era of Internet, many people choose to order food online, while the regulation of online food safety is faced with enormous challenges. Through the analysis of the comments data from third-party platform, a food safety evaluation dataset of violation and risk is constructed. In order to find the relationship between the comment data and risks level of online food, a novel parallelized distributed long and short term memory network model is proposed to predict the risk value of merchants, and an early warning system for network takeout merchants is established.