SimLex-999 (State of the art)

From ACL Wiki
Revision as of 12:43, 28 June 2015 by Pdturney (talk | contribs)
Jump to navigation Jump to search

SimLex-999 aims at a cleaner benchmark of similarity (but not relatedness). Pairs of words were chosen to represent different ranges of similarity and with either high or low association. Subjects were instructed to differentiate between similarity and relatedness and rate regarding the former only.

See also: Similarity (State of the art), Similar-Associated-Both Test Collection (State of the art).

Algorithm Reference for algorithm Reference for reported results Type Spearman's rho Pearson's r Notes
Neural language model Collobert & Weston (2008)[1] Hill et al. (2014a)[2] Distributional 0.268 - Trained on Wikipedia
Neural language model with global context Huang et al. (2012)[3] Hill et al. (2014a)[2] Distributional 0.098 - Trained on Wikipedia
Word2vec Mikolov et al. (2013)[4] Hill et al. (2014a)[2] Distributional 0.414 - Trained on Wikipedia
Lesk Banjade et al. (2015)[5] 0.404 0.347
UMBC Han et al. (2013)[6] Banjade et al. (2015)[5] 0.558 0.557 without using POS information
SVR4 Banjade et al. (2015)[5] Banjade et al. (2015)[5] Combined 0.642 0.658
ESA Banjade et al. (2015)[5] 0.271 0.145
RNNenc Hill et al. (2014b)[7] Hill et al. (2014b)[7] Distributional, multilingual 0.52 -

References

  1. ↑ R. Collobert and J. Weston. 2008. A unified architecture for natural language pro- cessing: Deep neural networks with multitask learning. In International Conference on Machine Learn- ing, ICML.
  2. ↑ 2.0 2.1 2.2 Hill, F., Reichart, R., & Korhonen, A. (2014a). SimLex-999: Evaluating Semantic Models with (Genuine) Similarity Estimation. Computation and Language.
  3. ↑ Eric H Huang, Richard Socher, Christopher D Manning, and Andrew Y Ng. 2012. Improving word representations via global context and multiple word prototypes. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1, pages 873–882. Association for Computational Linguistics.
  4. ↑ Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean. 2013. Efficient estimation of word representations in vector space. In Proceedings of International Conference of Learning Representations, Scottsdale, Arizona, USA.
  5. ↑ 5.0 5.1 5.2 5.3 5.4 Banjade, R., Maharjan, N., Niraula, N., Rus, V., & Gautam, D. (2015). Lemon and Tea Are Not Similar: Measuring Word-to-Word Similarity by Combining Different Methods. Computational Linguistics and Intelligent Text Processing, 9041, 335–346. doi:10.1007/978-3-319-18111-0_25
  6. ↑ Han, L., Kashyap, A., Finin, T., Mayfield, J., Weese, J.: UMBC EBIQUITY-CORE: Semantic textual similarity systems. In: Proceedings of the Second Joint Conference on Lexical and Computational Semantics, vol. 1, pp. 44–52 (2013)
  7. ↑ 7.0 7.1 Hill, F., Cho, K., Jean, S., Devin, C., & Bengio, Y. (2014b). Not All Neural Embeddings are Born Equal, 1–5.
Retrieved from "https://aclweb.org/aclwiki/index.php?title=SimLex-999_(State_of_the_art)&oldid=11127"