Prediction of Network Public Opinion Evolution Trends in Emergent Hot Events

Abstract

In recent years, there has been a notable increase in food safety incidents, which has raised considerable public concern. Optimizing food safety supervision and enhancing public trust have become urgent issues to be addressed. This study specifically examines the “tanker mixed with edible oil” incident and employs a variety of methodologies, including text analysis and time series modeling, to conduct a comprehensive analysis of public sentiment, The findings provide a scientific foundation for enhancing regulatory oversight. Relevant data were gathered via Python, public opinion trends were forecast via the ARIMA time series model, and an in-depth analysis of the thematic characteristics associated with each phase of public opinion development was conducted by integrating LDA topic modeling techniques. Meanwhile, this study employs social network analysis to construct an interactive network among users and identify key nodes and pathways involved in the dissemination of public opinion. Through simulation analysis, the following conclusions are drawn: (1) The “tanker mixed with cooking oil” incident exhibited a pronounced trend of negative sentiment that intensified over time. (2) The thematic analysis reveals public concern regarding disarray in food transportation and insufficient regulatory oversight, highlighting a shift in the public's focus. (3) Social network analysis emphasizes the crucial roles played by official media and individual key opinion leaders (KOLs) in shaping public opinion, illustrating how these entities influence the direction of public sentiment through their interactive relationships. Through the empirical analysis of the “tanker mixed with edible oil” incident, this paper verifies the effectiveness of the adopted method, providing an important reference for the risk prevention and control of food safety public opinion and policy-making.