Dialogue structure as a preference in anaphora resolution systems

2001

Journal of Computer Science, 2007

Anaphora resolution attempts to determine the correct antecedent of an anaphor (the term pointing back). In what follows, we propose an algorithm for the resolution of anaphoric pronouns that relies on lexical and syntactic knowledge incorporated in a modular approach based on constraints and preferences. Our objective was to find the correct antecedent to the following subject pronouns (il, ils, elle, elles), object pronouns (l', le, la, les) and possessive pronouns (son, sa, ses, leur, leurs) in unrestricted texts. We also identify and eliminate pleonastic pronouns and discard candidates appearing in appositions. Moreover we use a focus mechanism to determine salient entities. The algorithm, implemented in Prolog, realizes a success rate of 68%, which was considered a good performance for unrestricted French texts.

2008

Evaluation campaigns have become an established way to evaluate automatic systems which tackle the same task. This paper presents the first edition of the Anaphora Resolution Exercise (ARE) and the lessons learnt from it. This first edition focused only on English pronominal anaphora and NP coreference, and was organised as an exploratory exercise where various issues were investigated. ARE proposed four different tasks: pronominal anaphora resolution and NP coreference resolution on a predefined set of entities, pronominal anaphora resolution and NP coreference resolution on raw texts. For each of these tasks different inputs and evaluation metrics were prepared. This paper presents the four tasks, their input data and evaluation metrics used. Even though a large number of researchers in the field expressed their interest to participate, only three institutions took part in the formal evaluation. The paper briefly presents their results, but does not try to interpret them because in this edition of ARE our aim was not about finding why certain methods are better, but to prepare the ground for a fully-fledged edition.

Proceedings of the International Conference on Language Resources and Evaluation (LREC 2004), 2004

The paper presents an anaphora resolution algorithm for unrestricted text. In particular, we examine portability of a knowledge-based approach of (Mitamura et al., 2002), proposed for a domain-specific task. We obtain up to 70% accuracy on unrestricted text, which is a significant improvement (almost 20%) over a baseline we set for general text. As the overall results leave much room for improvement, we provide a detailed error analysis and investigate possible enhancements.

In this paper, we present a Paninian grammar based heuristic model 1 to resolve entity-pronoun references in Hindi dialogue. We explore the use of Paninian based dependency structures as a source of syntactico-semantic information. Our experiments illustrate that the use of dependency and dialogue structures help to resolve specific types of references. We also show that named entity, discourse information like subtopic boundary and animacy features increase the overall resolution accuracy to 64% for user-user interaction data and 59% for play-story corpora.

Proceedings of the 20th international conference on Computational Linguistics - COLING '04, 2004

This paper describes an extension of the dar-algorithm for resolving intersentential pronominal anaphors referring to individual and abstract entities in texts and dialogues. In dar individual entities are resolved combining models which identify high degree of salience with high degree of givenness (topicality) of entities in the hearer's cognitive model, e.g. ), with Hajičová et al.'s (1990 salience account which assigns the highest degree of salience to entities in the focal part of an utterance in Information Structure terms, which often introduce new information in discourse. Anaphors referring to abstract entities are resolved with an extension of the algorithm presented by . The extended dar-algorithm accounts for differences in the resolution mechanisms of different types of Danish pronouns. Manual tests of the algorithm show that dar performs better than other resolution algorithms on the same data.

This paper describes a standalone, publicly-available implementation of the Resolution of Anaphora Procedure (RAP) given by Lappin and Leass (1994). The RAP algorithm resolves third person pronouns, lexical anaphors, and identifies pleonastic pronouns. Our implementation, JavaRAP, fills a current need in anaphora resolution research by providing a reference implementation that can be benchmarked against current algorithms. The implementation uses the standard, publicly available Charniak (2000) parser as input, and generates a list of anaphora-antecedent pairs as output. Alternately, an in-place annotation or substitution of the anaphors with their antecedents can be produced. Evaluation on the MUC-6 co-reference task shows that JavaRAP has an accuracy of 57.9%, similar to the performance given previously in the literature (e.g., Preiss 2002).

International Journal of Applied Information Systems, 2013

Information management is an important requirement in today"s world. The anaphoric references hide the important information. The identification of anaphorically referred information is called as anaphora resolution which has significant impact on increasing the efficiency of information management, techniques including text summarization, information extraction etc. In this paper we have proposed a method for anaphora resolution to engineer information management. The proposed method acceptably determines the potential referents of the anaphora specially of the verb phrase form, distinguishes between pleonastic "it" and the anaphoric "it" and resolve the anaphora which is referred to after an interval of multiple sentences. The referents are stored in a list in the order of their occurrence in the discourse and eliminated from the list if they are not referred for to long. "Recency" is used as a salience factor to select the correct referent if other information like gender, number and type are not suffices to estimate the correct referent for an anaphor. To achieve a more precise resolution system WordNet lexical database is exploited to compare the synonyms of the anaphor with its possible referents.

International Journal in Foundations of Computer Science & Technology, 2014

Anaphora Resolution is the process of finding referents in the given discourse. Anaphora Resolution is one of the complex tasks of linguistics. This paper presents the performance analysis of two computational models that uses Gazetteer method for resolving anaphora in Hindi Language. In Gazetteer method different classes (Gazettes) of elements are created. These Gazettes are used to provide external knowledge to the system. The two models use Recency and Animistic factor for resolving anaphors. For Recency factor first model uses the concept of centering approach and second uses the concept of Lappin Leass approach. Gazetteers are used to provide Animistic knowledge. This paper presents the experimental results of both the models. These experiments are conducted on short Hindi stories, news articles and biography content from Wikipedia. The respective accuracy for both the model is analyzed and finally the conclusion is drawn for the best suitable model for Hindi Language.

… of the Sixth Workshop on Very …, 1998

This paper presents an algorithm for identi-fying pronominal anaphora and two experi-ments based upon this algorithm. We incorpo-rate multiple anaphora resolution factors into a statistical framework -- specifically the dis-tance between the pronoun and the ...