A Bootstrapping Approach to Unsupervised Detection of Cue Phrase Variants

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 921–928, Sydney, July 2006. c⃝2006 Association for Computational Linguistics A Bootstrapping Approach to Unsupervised Detection of Cue Phrase Variants Rashid M. Abdalla and Simone Teufel Computer Laboratory, University of Cambridge 15 JJ Thomson Avenue, Cambridge CB3 OFD, UK rma33@cam.ac.uk, sht25@cam.ac.uk Abstract We investigate the unsupervised detection of semi-fixed cue phrases such as “This paper proposes a novel approach. . . 1” from unseen text, on the basis of only a handful of seed cue phrases with the de- sired semantics. The problem, in contrast to bootstrapping approaches for Question Answering and Information Extraction, is that it is hard to find a constraining context for occurrences of semi-fixed cue phrases. Our method uses components of the cue phrase itself, rather than external con- text, to bootstrap. It successfully excludes phrases which are different from the tar- get semantics, but which look superficially similar. The method achieves 88% ac- curacy, outperforming standard bootstrap- ping approaches. 1 Introduction Cue phrases such as “This paper proposes a novel approach to. . . ”, “no method for . . . exists” or even “you will hear from my lawyer” are semi-fixed in that they constitute a formulaic pattern with a clear semantics, but with syntactic and lexical variations which are hard to predict and thus hard to detect in unseen text (e.g. “a new algorithm for . . . is suggested in the current paper” or “I envisage le- gal action”). In scientific discourse, such meta- discourse (Myers, 1992; Hyland, 1998) abounds and plays an important role in marking the dis- course structure of the texts. Finding these variants can be useful for many text understanding tasks because semi-fixed cue phrases act as linguistic markers indicating the im- portance and/or the rhetorical role of some ad- jacent text. For the summarisation of scientific 1In contrast to standard work in discourse linguistics, which mostly considers sentence connectives and adverbials as cue phrases, our definition includes longer phrases, some- times even entire sentences. papers, cue phrases such as “Our paper deals with. . . ” are commonly used as indicators of extraction-worthiness of sentences (Kupiec et al., 1995). Re-generative (rather than extractive) sum- marisation methods may want to go further than that and directly use the knowledge that a certain sentence contains the particular research aim of a paper, or a claimed gap in the literature. Similarly, in the task of automatic routing of customer emails and automatic answering of some of these, the de- tection of threats of legal action could be useful. However, systems that use cue phrases usually rely on manually compiled lists, the acquisition of which is time-consuming and error-prone and results in cue phrases which are genre-specific. Methods for finding cue phrases automatically in- clude Hovy and Lin (1998) (using the ratio of word frequency counts in summaries and their cor- responding texts), Teufel (1998) (using the most frequent n-grams), and Paice (1981) (using a …

Content truncated to 500 words