Informative and Controllable Opinion Summarization

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 2662–2672 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 2662 Informative and Controllable Opinion Summarization Reinald Kim Amplayo Mirella Lapata Institute for Language, Cognition and Computation School of Informatics, University of Edinburgh reinald.kim@ed.ac.uk, mlap@inf.ed.ac.uk Abstract Opinion summarization is the task of automat- ically generating summaries for a set of re- views about a specific target (e.g., a movie or a product). Since the number of reviews for each target can be prohibitively large, neural network-based methods follow a two-stage ap- proach where an extractive step first pre-selects a subset of salient opinions and an abstractive step creates the summary while conditioning on the extracted subset. However, the extrac- tive model leads to loss of information which may be useful depending on user needs. In this paper we propose a summarization frame- work that eliminates the need to rely only on pre-selected content and waste possibly use- ful information, especially when customizing summaries. The framework enables the use of all input reviews by first condensing them into multiple dense vectors which serve as input to an abstractive model. We showcase an effective instantiation of our framework which produces more informative summaries and also allows to take user preferences into account using our zero-shot customization technique. Experimen- tal results demonstrate that our model improves the state of the art on the Rotten Tomatoes dataset and generates customized summaries effectively. 1 Introduction The proliferation of opinions expressed in online reviews, blogs, and social media has created a press- ing need for automated systems which enable cus- tomers and companies to make informed decisions without having to absorb large amounts of opin- ionated text. Opinion summarization is the task of automatically generating summaries for a set of opinions about a specific target (Conrad et al., 2009). Figure 1 shows various reviews about the movie “Coach Carter” and example summaries gen- erated by humans and automatic systems. “Coach Carter” Reviews • Samuel L. Jackson plays the real-life coach of a high school basketball team in this solid sports drama ... • Great performance by Samuel Jackson but predictable as a slam dunk ... • ... excellent basketball choreography, Coach Carter is fun, hopeful, occasionally silly and, what can I say, inspiring. Consensus Summary Even though it’s based on a true story, Coach Carter is pretty formulaic stuff, but it’s effective and energetic, thanks to a strong central performance from Samuel L. Jackson. EXTRACT-ABSTRACT Framework Coach Carter is a preposterously plotted thriller that :::::: borrows ::::: heavily:::: from:::: other::::::: superior :::: films. (factually incorrect) CONDENSE-ABSTRACT Framework General: An inspirational flick with a healthy dose of mes- sage, but it’s too predictable. Customized (acting): An inspirational flick with a healthy dose of humor, Coach Carter is a perceptive sports drama with a standout performance from Samuel L. Jackson. Customized (plot): A feel-good tale with a healthy dose of heart, Coach Carter is a worthy addition to the basketball system …

Content truncated to 500 words