Mining Online Reviews For Predicting Sales Performance in Movie Domain

Nowadays people posting reviews are increasing and it gives us the knowledge to buy the products. People can express their sentiments through reviews posted in websites, In this paper, we takes the movie domain and by the use of the reviews we can predict the movie sales performance. The question is how the sentiment factor can be analyzed in the reviews? For the sentiment factor, we propose Sentiment PLSA(S-PLSA) which helps us to find the hidden sentiment factors in the reviews and helps us to classify the reviews. Then we have to predict the sales performance, for the prediction regression is the best way to do and we suggest ARSA, an Autoregressive Sentiment Aware Model. To improve the prediction accuracy to enhance the quality factor we also consider ARSQA, an Autoregressive Sentiment and Quality Aware model to build the quality for predicting sales performance.