Mining of Film and Television user Comments and Prediction of Film and Television Popularity based on Sentiment Analysis Algorithm

With the development of the Internet, the number of user comments on film and television works has increased sharply. These comments contain rich audience feelings and views, and are extremely valuable for understanding the popularity of film and television works. This article introduces sentiment analysis algorithms to effectively extract emotional information from massive comments, in order to predict the popularity of film and television works, improve the scientificity and accuracy of prediction, and apply sentiment analysis algorithms to determine user comment sentiment tendencies. Based on emotional information, a film and television popularity prediction model is constructed. Experiments have shown that in practical applications, the model can effectively predict the popularity of movies and TV shows, with an accuracy rate of up to 0.88 for sentiment classification comments. This study provides more accurate decision-making basis for the film and television industry.