Dynamic Influence of Network Public Opinions on Price Fluctuation of Small Agricultural Products Based on NLP-TVP-VAR Model—Taking Garlic as an Example

In recent years, the price of small agricultural products has both plummeted and skyrocketed, which a great impact on people’s lives. Studying factors affecting fluctuation is significance for stabilizing their price. With development application social media, farmers consumers are more greatly influenced by online public opinion, resulting in irrational planting behavior or purchasing behavior, complex products. Taking garlic as an example, we crawled through network opinions about from January 2015 to December 2020 using web crawler technology. Then, were quantified natural language processing time-varying parameter vector autoregression (NLP-TVP-VAR) model empirically analyze dynamic influence fluctuation. It was found that attitude attention have short-term fluctuation, influences each differ according direction, intensity timing. The positive, while largely negative. stronger than shows trend intensifying, been weaker previous years. addition, based results our study, present some recommendations improving comprehensive information platform early warning system whole industry chain