Kyoto: An integrated system for specific domain WSD

Autor: Soroa, A., Agirre, E., Lopez Lacalle, O., Bosma, W. E., Piek Vossen, Monachini, M., Lo, J., Kai Hsieh, K.
Přispěvatelé: Erk, K., Strapparava, C.
Předmět:
Zdroj: Scopus-Elsevier
SemeEval2010-5th International Workshop on Semantic Evaluation, pp. 417–420, Uppsala, Sweden, 15-16 Luglio 2010
info:cnr-pdr/source/autori:Soroa A.; Agirre E.; López De Lacalle O.; Bosma W.; Vossen P.; Monachini M.; Lo J.; Hsieh S./congresso_nome:SemeEval2010-5th International Workshop on Semantic Evaluation/congresso_luogo:Uppsala, Sweden/congresso_data:15-16 Luglio 2010/anno:2010/pagina_da:417/pagina_a:420/intervallo_pagine:417–420
Vrije Universiteit Amsterdam
Soroa, A, Agirre, E, Lopez de Lacalle, O, Bosma, W E, Vossen, P T J M, Monachini, M, Lo, J & Kai Hsieh, K 2010, Kyoto: An Integrated System for Specific Domain WSD . in K Erk & C Strapparava (eds), Proceedings of SemEval-2010: 5th International Workshop on Semantic Evaluations on Kyoto's subtask WSD17: All-words Word Sense Disambiguation on a Specific Domain, workshop collocation: ACL2010 . Association for Computational Linguistics (ACL), Uppsala, pp. 417-420, SemEval-2010: 5th International Workshop on Semantic Evaluations on Kyoto's subtask WSD17: All-words Word Sense Disambiguation on a Specific Domain, 11/07/10 . < http://aclweb.org/anthology-new/S/S10/S10-1093.pdf >
Popis: This document describes the preliminary release of the integrated Kyoto system for specific domain WSD. The system uses concept miners (Tybots) to extract domain-related terms and produces a domain-related thesaurus, followed by knowledge-based WSD based on wordnet graphs (UKB). The resulting system can be applied to any language with a lexical knowledge base, and is based on publicly available software and resources. Our participation in Semeval task #17 focused on producing running systems for all languages in the task, and we attained good results in all except Chinese. Due to the pressure of the time-constraints in the competition, the system is still under development, and we expect results to improve in the near future.
Databáze: OpenAIRE