Large-Scale and Multi-Perspective Opinion Summarization with Diverse Review Subsets
This addresses the need for scalable and multi-perspective opinion summarization in domains like e-commerce and entertainment, though it appears incremental as it builds on existing supervised methods with novel sampling and training components.
The paper tackles the problem of summarizing large sets of reviews from multiple perspectives by proposing SUBSUMM, a framework that selects diverse review subsets and uses a two-stage training scheme, resulting in effective generation of pros, cons, and verdict summaries from hundreds of reviews.
Opinion summarization is expected to digest larger review sets and provide summaries from different perspectives. However, most existing solutions are deficient in epitomizing extensive reviews and offering opinion summaries from various angles due to the lack of designs for information selection. To this end, we propose SUBSUMM, a supervised summarization framework for large-scale multi-perspective opinion summarization. SUBSUMM consists of a review sampling strategy set and a two-stage training scheme. The sampling strategies take sentiment orientation and contrastive information value into consideration, with which the review subsets from different perspectives and quality levels can be selected. Subsequently, the summarizer is encouraged to learn from the sub-optimal and optimal subsets successively in order to capitalize on the massive input. Experimental results on AmaSum and Rotten Tomatoes datasets demonstrate that SUBSUMM is adept at generating pros, cons, and verdict summaries from hundreds of input reviews. Furthermore, our in-depth analysis verifies that the advanced selection of review subsets and the two-stage training scheme are vital to boosting the summarization performance.