Automatic Aggregation by Joint Modeling of Aspects and Values
The authors present a model for aggregation of product review snippets by joint aspect identification and sentiment analysis. Their model simultaneously identifies an underlying set of ratable aspects presented in the reviews of a product (e.g., sushi and miso for a Japanese restaurant) and determines the corresponding sentiment of each aspect. This approach directly enables discovery of highly-rated or inconsistent aspects of a product. Their generative model admits an efficient variational mean-field inference algorithm. It is also easily extensible and they describe several modifications and their effects on model structure and inference. They test their model on two tasks, joint aspect identification and sentiment analysis on a set of Yelp reviews and aspect identification alone on a set of medical summaries.