Пожалуйста, используйте этот идентификатор, чтобы цитировать или ссылаться на этот ресурс: http://elar.urfu.ru/handle/10995/89994
Название: Unsupervised, knowledge-free, and interpretable word sense disambiguation
Авторы: Panchenko, A.
Marten, F.
Ruppert, E.
Faralli, S.
Ustalov, D.
Ponzetto, S. P.
Biemann, C.
Дата публикации: 2017
Издатель: Association for Computational Linguistics (ACL)
Библиографическое описание: Unsupervised, knowledge-free, and interpretable word sense disambiguation / A. Panchenko, F. Marten, E. Ruppert, S. Faralli, et al. . — DOI 10.18653/v1/d17-2016 // EMNLP 2017 - Conference on Empirical Methods in Natural Language Processing: System Demonstrations, Proceedings. — 2017. — P. 91-96.
Аннотация: Interpretability of a predictive model is a powerful feature that gains the trust of users in the correctness of the predictions. In word sense disambiguation (WSD), knowledge-based systems tend to be much more interpretable than knowledge-free counterparts as they rely on the wealth of manually-encoded elements representing word senses, such as hypernyms, usage examples, and images. We present a WSD system that bridges the gap between these two so far disconnected groups of methods. Namely, our system, providing access to several state-of-the-art WSD models, aims to be interpretable as a knowledge-based system while it remains completely unsupervised and knowledge-free. The presented tool features a Web interface for all-word disambiguation of texts that makes the sense predictions human readable by providing interpretable word sense inventories, sense representations, and disambiguation results. We provide a public API, enabling seamless integration. © 2017 Association for Computational Linguistics.
Ключевые слова: KNOWLEDGE BASED SYSTEMS
HUMAN-READABLE
INTERPRETABILITY
PREDICTIVE MODELING
SEAMLESS INTEGRATION
STATE OF THE ART
WEB INTERFACE
WORD SENSE
WORD-SENSE DISAMBIGUATION
NATURAL LANGUAGE PROCESSING SYSTEMS
URI: http://elar.urfu.ru/handle/10995/89994
Условия доступа: info:eu-repo/semantics/openAccess
cc-by
Идентификатор SCOPUS: 85072911197
Идентификатор PURE: 11097509
ISBN: 9781945626975
DOI: 10.18653/v1/d17-2016
Сведения о поддержке: Deutsche Forschungsgemeinschaft, DFG
Russian Foundation for Basic Research, RFBR: 16-37-00354
Amazon Web Services, AWS
Microsoft
We acknowledge the support of the DFG under the “JOIN-T” project, the RFBR under project no. 16-37-00354 mol a, Amazon via the “AWS Research Grants” and Microsoft via the “Azure for Research” programs. Finally, we also thank four anonymous reviewers for their helpful comments.
Располагается в коллекциях:Научные публикации ученых УрФУ, проиндексированные в SCOPUS и WoS CC

Файлы этого ресурса:
Файл Описание РазмерФормат 
10.18653-v1-d17-2016.pdf864,76 kBAdobe PDFПросмотреть/Открыть


Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.