Reference Scope Identification in Citing Sentences

2012 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 80–90, Montr´eal, Canada, June 3-8, 2012. c⃝2012 Association for Computational Linguistics Reference Scope Identification in Citing Sentences Amjad Abu-Jbara EECS Department University of Michigan Ann Arbor, MI, USA amjbara@umich.edu Dragomir Radev EECS Department University of Michigan Ann Arbor, MI, USA radev@umich.edu Abstract A citing sentence is one that appears in a sci- entific article and cites previous work. Cit- ing sentences have been studied and used in many applications. For example, they have been used in scientific paper summarization, automatic survey generation, paraphrase iden- tification, and citation function classification. Citing sentences that cite multiple papers are common in scientific writing. This observa- tion should be taken into consideration when using citing sentences in applications. For in- stance, when a citing sentence is used in a summary of a scientific paper, only the frag- ments of the sentence that are relevant to the summarized paper should be included in the summary. In this paper, we present and com- pare three different approaches for identifying the fragments of a citing sentence that are re- lated to a given target reference. Our methods are: word classification, sequence labeling, and segment classification. Our experiments show that segment classification achieves the best results. 1 Introduction Citation plays an important role in science. It makes the accumulation of knowledge possible. When a reference appears in a scientific article, it is usually accompanied by a span of text that highlights the important contributions of the cited article. We call a sentence that contains an explicit reference to previous work a citation sentence. For example, sentence (1) below is a citing sentence that cites a paper by Philip Resnik and describes the problem Resnik addressed in his paper. (1) Resnik (1999) addressed the issue of language identification for finding Web pages in the languages of interest. Previous work has studied and used citation sen- tences in various applications such as: scientific pa- per summarization (Elkiss et al., 2008; Qazvinian and Radev, 2008; Mei and Zhai, 2008; Qazvinian et al., 2010; Qazvinian and Radev, 2010; Abu- Jbara and Radev, 2011), automatic survey genera- tion (Nanba et al., 2000; Mohammad et al., 2009), citation function classification (Nanba et al., 2000; Teufel et al., 2006; Siddharthan and Teufel, 2007; Teufel, 2007), and paraphrase recognition (Nakov et al., 2004; Schwartz et al., 2007). Sentence (1) above contains one reference, and the whole sentence is talking about that reference. This is not always the case in scientific writing. Sentences that contain references to multiple papers are very common. For example, sentence (2) below contains three references. (2) Grefenstette and Nioche (2000) and Jones and Ghani (2000) use the web to generate corpora for languages where electronic resources are scarce, while Resnik (1999) describes a method for mining the web for bilingual texts. 80 The first fragment describes the contribution of Grefenstette and Nioche (2000) and Jones and Ghani (2000). The second fragment describes the contribu- tion …

Content truncated to 500 words