Efficient Opinion Summarization on Comments with Online-LDA

Customer reviews and comments on web pages are important information in our daily life. For example, we prefer to choose a hotel with positive comments from previous customers. As the huge amounts of such information demonstrate the characteristics of big data, it places heavy burdens on the assimilation of the customer-contributed opinions. To overcoming this problem, we study an efficient opinion summarization approach for a set of massive user reviews and comments associated with an online resource, to summarize the opinions into two categories, i.e., positive and negative. In this paper, we proposed a framework including: (1) overcoming the big data problem of online comments using the efficient online-LDA approach; (2) selecting meaningful topics from the imbalanced data; (3) summarizing the opinion of comments with high precision and recall. This framework is different from much of the previous work in that the topics are pre-defined and selected the topics for better opinion summarization. To evaluate the proposed framework, we perform the experiments on a dataset of hotel reviews for the variety of topics contained. The results show that our framework can gain a significant performance improvement on opinion summarization.