Unsupervised Ontology- And Sentiment-Aware Review Summarization

In this Web 2.0 era, there is an ever increasing number of customer reviews, which must be summarized to help consumers effortlessly make informed decisions. Previous work on reviews summarization has simplified the problem by assuming that aspects (e.g., "display") are independent of each other and that the opinion for each aspect in a review is Boolean: positive or negative. However, in reality aspects may be interrelated - e.g., "display" and "display color" - and the sentiment takes values in a continuous range - e.g., somewhat vs very positive. We present a novel, unsupervised review summarization framework that advances the state-of-the-art by leveraging a domain hierarchy of concepts to handle the semantic overlap among the aspects, and by accounting for different sentiment levels. We show that the problem is NP-hard and present bounded approximate algorithms to compute the most representative set of sentences or reviews, based on a principled opinion coverage framework. We experimentally evaluate the proposed algorithms on real datasets in terms of their efficiency and effectiveness compared to the optimal algorithms. We also show that our methods generate summaries of superior quality than several baselines in short execution times.