AI helps you reading Science
AI generates interpretation videos
AI extracts and analyses the key points of the paper to generate videos automatically
AI parses the academic lineage of this thesis
AI extracts a summary of this paper
Connecting Language and Knowledge Bases with Embedding Models for Relation Extraction.
empirical methods in natural language processing, (2013): 1366-1371
- Information extraction (IE) aims at generating structured data from free text in order to populate Knowledge Bases (KBs).
- One is given an incomplete KB composed of a set of triples of the form (h, r , t); h is the left-hand side entity, t the right-hand side entity and r the relationship linking them.
- This paper focuses on the problem of learning to perform relation extraction (RE) under weak supervision from a KB.
- RE is sub-task of IE that considers that entities have already been detected by a different process, such as a named-entity recognizer.
- RE aims at assigning to a relation mention m
- Information extraction (IE) aims at generating structured data from free text in order to populate Knowledge Bases (KBs)
- One is given an incomplete KB composed of a set of triples of the form (h, r, t); h is the left-hand side entity, t the right-hand side entity and r the relationship linking them
- This paper focuses on the problem of learning to perform relation extraction (RE) under weak supervision from a KB
- Figure 1 displays the aggregate precision / recall curves of our approach WSABIEM2R+FB which uses the combination of Sm2r + Skb, as well as WSABIEM2R, which only uses Sm2r, and existing state-of-the-art approaches: HOFFMANN (Hoffmann et al, 2011)2, MIMLRE (Surdeanu et al, 2012)
- In this paper we described a framework for leveraging large scale knowledge bases to improve relation extraction by training on pairs but using all other KB triples as well
- We empirically showed that it allows to significantly improve precision on extracted relations
- The addition of extra knowledge from other Freebase entities in WSABIEM2R+FB provides superior performance to all other methods, by a wide margin, at least between 0 and 0.1 recall.
- WSABIEM2R+FB for recall > 0.1 degrades rapidly, faster than that of other methods.
- This is caused by the simplifications of WSABIEM2R that prevent it from reaching high precision when the recall is greater than 0.1.
- The authors recall that Freebase data is not used to detect relationships i.e. to discriminate between NA and the rest, but only to select the best relationship in case of detection
- In this paper the authors described a framework for leveraging large scale knowledge bases to improve relation extraction by training on pairs but using all other KB triples as well.
- The authors' modeling approach is general and should apply to other settings, e.g. for the task of entity linking
- This work was carried out in the framework of the Labex MS2T (ANR-11-IDEX-0004-02), and funded by the French National Agency for Research (EVEREST-12-JS02-005-01)
- Michele Banko, Michael J Cafarella, Stephen Soderland, Matthew Broadhead, and Oren Etzioni. 2007. Open information extraction from the web. In IJCAI, volume 7, pages 2670–2676.
- Antoine Bordes, Nicolas Usunier, and Jason Weston. 2010. Label ranking under ambiguous supervision for learning semantic correspondences. In Proceedings of the 27th International Conference on Machine Learning (ICML-10), pages 103–110.
- Antoine Bordes, Jason Weston, Ronan Collobert, and Yoshua Bengio. 2011. Learning structured embeddings of knowledge bases. In Proc. of the 25th Conf. on Artif. Intel. (AAAI).
- Antoine Bordes, Xavier Glorot, Jason Weston, and Yoshua Bengio. 2012. Joint learning of words and meaning representations for open-text semantic parsing. In Proc. of the 15th Intern. Conf. on Artif. Intel. and Stat., volume 22, pages 127–135. JMLR.
- Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, and Oksana Yakhnenko. 2013. Translating embeddings for modeling multi-relational data. In Advances in Neural Information Processing Systems (NIPS 26).
- Mark Craven, Johan Kumlien, et al. 1999. Constructing biological knowledge bases by extracting information from text sources. In ISMB, volume 1999, pages 77– 86.
- Jenny Rose Finkel, Trond Grenager, and Christopher Manning. 2005. Incorporating non-local information into information extraction systems by gibbs sampling. In Proceedings of the 43rd Annual Meeting on Association for Computational Linguistics, pages 363–370. Association for Computational Linguistics.
- Raphael Hoffmann, Congle Zhang, Xiao Ling, Luke Zettlemoyer, and Daniel S Weld. 2011. Knowledgebased weak supervision for information extraction of overlapping relations. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, volume 1, pages 541–550.
- Rohit J Kate and Raymond J Mooney. 2007. Learning language semantics from ambiguous supervision. In AAAI, volume 7, pages 895–900.
- Ni Lao, Amarnag Subramanya, Fernando Pereira, and William W Cohen. 2012. Reading the web with learned syntactic-semantic inference rules. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 1017– 1026. Association for Computational Linguistics.
- Percy Liang, Michael I Jordan, and Dan Klein. 2009. Learning semantic correspondences with less supervision. In Proceedings of the Joint Conference of the