论文信息 - Semantic Relations for the Recognition of Anaphoric Expressions with dierent head

Semantic Relations for the Recognition of Anaphoric Expressions with dierent head

This technical report addresses the problem of how semantic relations can be used to improve the performance of anaphora resolution. In particular I took the mentions from coreferce chains classied as false negatives by BART, a system for anaphora resolution, and I extracted the relations that hold between the two dierent mentions by means of pattern searching in google. The results show that it is possible to improve the performance of anaphora resolution by classifying false negative co-referring mentions with the same head using taxonomic and associative relations as features.

Fabio Celli

[1] Yannick Versley,et al. BART: A Modular Toolkit for Coreference Resolution , 2008, ACL.

[2] Ian H. Witten,et al. Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[3] Marti A. Hearst. Automatic Acquisition of Hyponyms from Large Text Corpora , 1992, COLING.

[4] Pat Langley,et al. Estimating Continuous Distributions in Bayesian Classifiers , 1995, UAI.

[5] Massimo Poesio,et al. Extracting concept descriptions from the Web: the importance of attributes and values , 2008, Ontology Learning and Population.

[6] Dan I. Moldovan,et al. Automatic Discovery of Part-Whole Relations , 2006, CL.

[7] J. Ross Quinlan,et al. C4.5: Programs for Machine Learning , 1992 .

[8] Ian H. Witten,et al. Data mining: practical machine learning tools and techniques with Java implementations , 2002, SGMD.

[9] Ian Witten,et al. Data Mining , 2000 .