Corpus-based anaphora resolution towards antecedent preference

Paul, Michael; Yamamoto, Kazumasa; Sumita, Eiichiro

doi:10.3115/1608810.1608820

Cited by 11 publications

(6 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Knowledge-independent Approach [15], Machine Learning Approach [16] etc are also using the provision of corpus in their methods. The method proposed by Michael Paul in 2002 is an example of corpus based approach by combining a machine learning method and statistical information [17]. First, a decision tree trained on an annotated corpus determines the coreference relation of a given anaphor and antecedent candidates and is utilized as a filter in order to reduce the number of potential candidates.…”

Section: Existing Approaches -A Brief Notementioning

confidence: 99%

Anaphora Resolution using ThemeSets

Wilson¹,

Jeba²

2019

IJITEE

View full text Add to dashboard Cite

Anaphora resolution is a procedure of replacing pronouns with its referring nouns which may be available in the same sentence or in different sentences within the same document. Even multiple approaches are available for anaphora resolution; there may be some space for semantic approaches which provides solution better than syntactic or corpus based approaches. In syntactic approaches, all noun phrases from the previous or current sentences are checked for getting constrains agreement with the anaphor and score value is calculated for a pair of Noun-Pronoun based on these constrains. The pair getting the highest positive score is treated as the result. In some cases the sentence structure is not changed, hence the score corresponding to the feature values and syntactic structures are same. Here the actual replacement of pronoun with noun depends on the meaning of the sentence especially the meaning of the verb. To resolve such situations, here we are proposing a method which is a combination of syntactic and semantic approaches based on ThemeSets or thematic sets. Using ThemeSets we are exploiting the role of verbal lexemes associated with the noun or pronoun for the resolution of anaphora. Anaphora resolution in semantic way has great importance in the modern era of artificial intelligence which enhance multi-dimensional research in the area of natural language processing in a better way.

show abstract

Section: Existing Approaches -A Brief Notementioning

confidence: 99%

Anaphora Resolution using ThemeSets

Wilson¹,

Jeba²

2019

IJITEE

View full text Add to dashboard Cite

show abstract

“…Anaphora resolution research has been attracted intensive attentions from NLP community in the last few decades, and approaches to the anaphora resolution problem also vary. Earlier research efforts had focused on traditional heuristic-based approaches that explored various types of heuristics, typically involving the use of lexical, syntactic and semantic information [31,35], computational theories of discourse such as focusing [20] and centering algorithm [50,54], and statistics/corpus approach based on co-occurrence patterns [10,41]. In recent, supervised machine learning approaches have been widely explored in reference resolution and achieved considerable success.…”

Section: Related Workmentioning

confidence: 99%

Analysing anaphoric ambiguity in natural language requirements

et al. 2011

View full text Add to dashboard Cite

Many requirements documents are written in natural language (NL). However, with the flexibility of NL comes the risk of introducing unwanted ambiguities in the requirements and misunderstandings between stakeholders. In this paper, we describe an automated approach to identify potentially nocuous ambiguity, which occurs when text is interpreted differently by different readers. We concentrate on anaphoric ambiguity, which occurs when readers may disagree on how pronouns should be interpreted.We describe a number of heuristics, each of which captures information that may lead a reader to favour a particular interpretation of the text. We use these heuristics to build a classifier, which in turn predicts the degree to which particular interpretations are preferred. We collected multiple human judgements on the interpretation of requirements exhibiting anaphoric ambiguity, and show how the distribution of these judgements can be used to assess whether a particular instance of ambiguity is nocuous. Given a requirements document written in natural language, our approach can identify sentences which contain anaphoric ambiguity, and use the classifier to alert the requirements writer of text that runs the risk of misinterpretation.We report on a series of experiments that we conducted to evaluate the performance of the automated system we developed to support our approach. The results show that the system achieves high recall with a consistent improvement on baseline precision subject to some ambiguity tolerance levels, allowing us to explore and highlight realistic and potentially problematic ambiguities in actual requirements documents.

show abstract

“…E.g., Dagan & Itai (1991) presented a corpus-based approach to the resolution of the pronoun it, but they use a written text corpus and do not mention non-NP-antecedents at all. Paul et al (1999) presented a corpus-based anaphora resolution algorithm for spoken dialogue. For their experiments, however, they restricted anaphoric relations to those with NP-antecedents.…”

Section: Comparison To Related Workmentioning

confidence: 99%