Constructing the Phrase Dictionary and Visualizing Consumer Behaviors in the Food Industry Based on Online Reviews during the COVID-19 Pandemic

Abstract

The COVID-19 pandemic affects food industrylargely. In this study, the data of online reviewsare collected from dianping.com during the outbreak and stable period of the COVID-19 pandemic in China, and the rules in combination with the statistical methods are adopted to train the dictionary ofrestaurant phrases.After the K-means algorithm is adopted to cluster the phrases in the dictionary, and the cluster class tags are defined, the co-occurrence analysis and the wordcloud analysis are conducted on the reviews. As indicated from the results, consumers pay attention to the three basic elements (i.e., services, environments and tastes), as well as to the social distance between people; Consumers who are more concerned about the pandemic situation raise higher requirements on environmental health issues than ordinary consumers, and place stress on the acquisition of security. As revealed from the mentioned results, restaurants should primarily take measures to maintain safe social distance between people and raise more rigorous environmental hygiene requirements on the environment. This method is served as a reference for the further online reviews analysis and provides implications for the management of the restaurants in COVID-19 pandemic period.