Document Summarization via Guided Sentence Compression

Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing, pages 490–500, Seattle, Washington, USA, 18-21 October 2013. c⃝2013 Association for Computational Linguistics Document Summarization via Guided Sentence Compression Chen Li1, Fei Liu2, Fuliang Weng2, Yang Liu1 1 Computer Science Department, The University of Texas at Dallas Richardson, Texas 75080, USA 2 Research and Technology Center, Robert Bosch LLC Palo Alto, California 94304, USA {chenli,yangl@hlt.utdallas.edu} {fei.liu, fuliang.weng@us.bosch.com} Abstract Joint compression and summarization has been used recently to generate high quality summaries. However, such word-based joint optimization is computationally expensive. In this paper we adopt the ‘sentence compression + sentence selection’ pipeline approach for compressive summarization, but propose to perform summary guided compression, rather than generic sentence-based compression. To create an annotated corpus, the human anno- tators were asked to compress sentences while explicitly given the important summary words in the sentences. Using this corpus, we train a supervised sentence compression model us- ing a set of word-, syntax-, and document- level features. During summarization, we use multiple compressed sentences in the inte- ger linear programming framework to select salient summary sentences. Our results on the TAC 2008 and 2011 summarization data sets show that by incorporating the guided sen- tence compression model, our summarization system can yield significant performance gain as compared to the state-of-the-art. 1 Introduction Automatic summarization can be broadly divided into two categories: extractive and abstractive sum- marization. Extractive summarization focuses on selecting the salient sentences from the document collection and concatenating them to form a sum- mary; while abstractive summarization is generally considered more difficult, involving sophisticated techniques for meaning representation, content plan- ning, surface realization, etc., and the “true abstrac- tive summarization remains a researcher’s dream” (Radev et al., 2002). There has been a surge of interest in recent years on generating compressed document sum- maries as a viable step towards abstractive sum- marization. These compressive summaries often contain more information than sentence-based ex- tractive summaries since they can remove insignif- icant sentence constituents and make space for more salient information that is otherwise dropped due to the summary length constraint. Two general strate- gies have been used for compressive summarization. One is a pipeline approach, where sentence-based extractive summarization is followed or proceeded by sentence compression (Knight and Marcu, 2000; Lin, 2003; Zajic et al., 2007; Wang et al., 2013). Another line of work uses joint compression and summarization. They have been shown to achieve promising performance (Daum´e, 2006; Martins and Smith, 2009; Berg-Kirkpatrick et al., 2011; Chali and Hasan, 2012; Almeida and Martins, 2013; Qian and Liu, 2013). One popular approach for such joint compression and summarization is via integer lin- ear programming (ILP). However, since words are the units in the optimization framework, solving this ILP problem can be expensive. In this study, we use the pipeline compression and summarization method because of its compu- tational efficiency. Prior work using such pipeline methods simply uses generic sentence-based com- pression for each sentence in the documents, no mat- ter whether compression …

Content truncated to 500 words