Improving Document-Level Sentiment Analysis with User and Product Context
This work provides an incremental improvement to document-level sentiment analysis for e-commerce and review platforms by leveraging additional contextual information.
This paper addresses the limitation of document-level sentiment analysis models that only consider the current review text when incorporating user and product information. By integrating historical reviews from the same user and about the same product, the authors achieved over 2 percentage points improvement on state-of-the-art across IMDB, Yelp 2013, and Yelp 2014 datasets.
Past work that improves document-level sentiment analysis by encoding user and product information has been limited to considering only the text of the current review. We investigate incorporating additional review text available at the time of sentiment prediction that may prove meaningful for guiding prediction. Firstly, we incorporate all available historical review text belonging to the author of the review in question. Secondly, we investigate the inclusion of historical reviews associated with the current product (written by other users). We achieve this by explicitly storing representations of reviews written by the same user and about the same product and force the model to memorize all reviews for one particular user and product. Additionally, we drop the hierarchical architecture used in previous work to enable words in the text to directly attend to each other. Experiment results on IMDB, Yelp 2013 and Yelp 2014 datasets show improvement to state-of-the-art of more than 2 percentage points in the best case.