Abstractive Multi-document Summarization with Semantic Information Extraction

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 1908–1913, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. Abstractive Multi-document Summarization with Semantic Infor- mation Extraction Wei Li Key Lab of Intelligent Info. Processing, Institute of Computing Technology, CAS Beijing, 100190, China weili@kg.ict.ac.cn Abstract This paper proposes a novel approach to generate abstractive summary for multi- ple documents by extracting semantic in- formation from texts. The concept of Basic Semantic Unit (BSU) is defined to describe the semantics of an event or ac- tion. A semantic link network on BSUs is constructed to capture the semantic in- formation of texts. Summary structure is planned with sentences generated based on the semantic link network. Experi- ments demonstrate that the approach is effective in generating informative, co- herent and compact summary. 1 Introduction Most automatic summarization approaches are extractive which leverage only literal or syntactic information in documents. Sentences are extract- ed from the original documents directly by rank- ing or scoring and only little post-editing is made (Yih et al., 2007; Wan et al., 2007; Wang et al., 2008; Wan and Xiao, 2009). Pure extraction has intrinsic limits compared to abstraction (Carenini and Cheung, 2008). Abstractive summarization requires semantic analysis and abstract representation of texts, which need knowledge on and beyond the texts (Zhuge, 2015a). There are some abstractive ap- proaches in recent years: sentence compression (Knight and Marcu, 2000; Knight and Marcu, 2002; Cohn and Lapata, 2009), sentence fusion (Barzilay and McKeown, 2005; Filippova and Strube, 2008), and sentence revision (Tanaka et al., 2009). However, these approaches are sen- tence rewriting techniques based on syntactical analysis without semantic analysis and abstract representation. Fully abstractive summarization approach re- quires a separate process for the analysis of texts that serves as an intermediate step before the generation of sentences (Genest and Lapalme, 2011). Statistics of words or phrases and syntac- tical analysis that have been widely used in exist- ing summarization approaches are all shallow processing of text. It is necessary to explore summarization methods based on deeper seman- tic analysis. We define the concept of Basic Semantic Unit (BSU) to express the semantics of texts. A BSU is an action indicator with its obligatory argu- ments which contain actor and receiver of the action. BSU is the most basic element of coher- ent information in texts, which can describe the semantics of an event or action. The semantic information of texts is represented by extracting BSUs and constructing BSU semantic link net- work (Zhuge, 2009). Semantic Link Network consists of semantic nodes, semantic links and reasoning rules (Zhuge, 2010; 2011; 2012; 2015b). The semantic nodes can be any resources. In this work, the semantic nodes are BSUs ex- tracted from texts. We use semantic relatedness between BSUs as semantic links. Then summary can be generated based on the semantic link net- work through summary structure planning. The characteristics of our approaches are as follows:  Each BSU describes the semantics of an event or action. …

Content truncated to 500 words