The Instantiation Discourse Relation: A Corpus Analysis of Its Properties and Improved Detection
23029Authors
nan
Abstract
INSTANTIATION is a fairly common discourse relation and past work has suggested that it plays special roles in local coherence, in sen- timent expression and in content selection in summarization. In this paper we provide the first systematic corpus analysis of the rela- tion and show that relation-specific features can improve considerably the detection of the relation. We show that sentences involved in INSTANTIATION are set apart from other sentences by the use of gradable (subjec- tive) adjectives, the occurrence of rare words and by different patterns in part-of-speech us- age. Words across arguments of INSTANTI- ATION are connected through hypernym and meronym relations significantly more often than in other sentences and that they stand out in context by being significantly less sim- ilar to each other than other adjacent sentence pairs. These factors provide substantial pre- dictive power that improves the identification of implicit INSTANTIATION relation by more than 5% F-measure. 1
Keywords
Full Content
Proceedings of NAACL-HLT 2016, pages 1181–1186, San Diego, California, June 12-17, 2016. c⃝2016 Association for Computational Linguistics The Instantiation Discourse Relation: A Corpus Analysis of Its Properties and Improved Detection Junyi Jessy Li and Ani Nenkova University of Pennsylvania {ljunyi,nenkova}@seas.upenn.edu Abstract INSTANTIATION is a fairly common discourse relation and past work has suggested that it plays special roles in local coherence, in sen- timent expression and in content selection in summarization. In this paper we provide the first systematic corpus analysis of the rela- tion and show that relation-specific features can improve considerably the detection of the relation. We show that sentences involved in INSTANTIATION are set apart from other sentences by the use of gradable (subjec- tive) adjectives, the occurrence of rare words and by different patterns in part-of-speech us- age. Words across arguments of INSTANTI- ATION are connected through hypernym and meronym relations significantly more often than in other sentences and that they stand out in context by being significantly less sim- ilar to each other than other adjacent sentence pairs. These factors provide substantial pre- dictive power that improves the identification of implicit INSTANTIATION relation by more than 5% F-measure. 1 Introduction In an INSTANTIATION relation, one text span ex- plains in further detail the events, reasons, behaviors and attitudes mentioned in the other (Miltsakaki et al., 2008), as illustrated by the segments below: [a] Other fundamental “reforms” of the 1986 act have been threatened as well. [b] The House seriously considered raising the top tax rate paid by individuals with the highest incomes. Sentence [a] mentions “other reforms” and a threat to them, but leaves unspecified what are the reforms or how they are threatened. Sentence [b] provides sufficient detail for the reader to infer more con- cretely what has happened. The INSTANTIATION relation has some special properties. A study of discourse relations as in- dicators for content selection in single document summarization revealed that the first sentences from INSTANTIATION pairs are included in human sum- maries significantly more often than other sentences (Louis et al., 2010) and that being a first sentence in an INSTANTIATION relation is the most powerful indicator for content selection related to discourse relation sense. The sentences between which the relation holds also contain more sentiment expres- sions than other sentences (Trnavac and Taboada, 2013), making it a special target for sentiment anal- ysis applications. Moreover, INSTANTIATION rela- tions appear to play a special role in local coherence (Louis and Nenkova, 2010), as the flow between IN- STANTIATION sentences is not explained by the ma- jor coherence theories (Kehler, 2004; Grosz et al., 1995). Many of the sentences in INSTANTIATION relation contain entity instantiations (complex ex- amples of set-instance anaphora), such as “several EU countries”—“the UK”, “footballers”—“Wayne Rooney” and “most cosmetic purchase”—“lipstick” (McKinlay and Markert, 2011), raising further ques- tions about the relationship between INSTANTIA- TIONS and key discourse phenomena. Detecting an INSTANTIATION, however, is hard. In the Penn Discourse Treebank (PDTB) (Prasad et al., 2008), INSTANTIATION is one of the …
Content truncated to 500 words
Labeled Text
INSTANTIATION is a fairly common discourse relation and past work has suggested that it plays special roles in local coherence, in sentiment expression and in content selection in summarization. In this paper we provide the first systematic corpus analysis of the relation and show that relation-specific features can improve considerably the detection of the relation. We show that sentences involved in INSTANTIATION are set apart from other sentences by the use of gradable (subjective) adjectives, the occurrence of rare words and by different patterns in part-of-speech usage. Words across arguments of INSTANTIATION are connected through hypernym and meronym relations significantly more often than in other sentences and that they stand out in context by being significantly less similar to each other than other adjacent sentence pairs. These factors provide substantial predictive power that improves the identification of implicit INSTANTIATION relation by more than 5% F-measure. 1
Publication Details
- Paper ID:
- 23029
- Year:
- 2016
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan