The meat market behavior has been studied typically based on the price and income effect. Recently, unstructured data have been generated from big data analysis. Therefore, our study generates a sentiment index about African Swine Fever(ASF) and estimates the impact of the change in the ASF sentiment index on the Korean meat market. Word2Vec technique, one of the Neural Network Language Model(NNLM), is used to generate the ASF sentiment index. Moreover, we decompose the ASF sentiment index into its positive and negative changes. Using this ASF sentiment index, we build a non-linear Autoregressive Distributed Lag(ARDL) to estimate how the positive and negative ASF sentiment index sensitively causes pork, beef, and chicken prices. The results show that pork and beef prices are more sensitive to the positive ASF sentiment index changes than the negative ASF sentiment index. Our study is the first trial to study the meat market involving sentiment analysis to the best of our knowledge. We expect that developed sentiment analysis will generate animal disease sentiment and help understand the response of animal disease sentiment in the Korean meat market.