Difference between revisions of "WordSimilarity-353 Test Collection (State of the art)"
Jump to navigation
Jump to search
| Line 43: | Line 43: | ||
* [[ESL Synonym Questions (State of the art)|ESL Synonym Questions]] | * [[ESL Synonym Questions (State of the art)|ESL Synonym Questions]] | ||
* [[SAT Analogy Questions]] | * [[SAT Analogy Questions]] | ||
| + | * [[TOEFL Synonym Questions (State of the art)|TOEFL Synonym Questions]] | ||
* [[State of the art]] | * [[State of the art]] | ||
[[Category:State of the art]] | [[Category:State of the art]] | ||
Revision as of 12:40, 3 October 2013
- WordSimilarity-353 Test Collection
- contains two sets of English word pairs along with human-assigned similarity judgements
- first set (set1) contains 153 word pairs along with their similarity scores assigned by 13 subjects
- second set (set2) contains 200 word pairs with similarity assessed by 16 subjects
- WordSimilarity-353 dataset is available here
- performance is measured by Spearman's rank correlation coefficient
- introduced by Finkelstein et al. (2002)
- subsequently used by many other researchers
0.75 Spearman rank correlation
http://www.cs.technion.ac.il/~gabr/papers/ijcai-2007-sim.pdf
0.65 Spearman rank correlation http://nlp.stanford.edu/~lmthang/data/papers/conll13_morpho.pdf
Table of results
| Algorithm | Reference | Spearman's rho |
|---|---|---|
| HSMN+csmRNN | Luong et al. (2013) | 0.646 |
References
Luong, Minh-Thang, Richard Socher, and Christopher D. Manning. (2013) Better word representations with recursive neural networks for morphology. CoNLL-2013: 104.