Difference between revisions of "ESL Synonym Questions (State of the art)"
Jump to navigation
Jump to search
| (19 intermediate revisions by 3 users not shown) | |||
| Line 2: | Line 2: | ||
* 50 multiple-choice synonym questions; 4 choices per question | * 50 multiple-choice synonym questions; 4 choices per question | ||
* each question includes a sentence, providing context for the question | * each question includes a sentence, providing context for the question | ||
| − | * ESL questions available from [http://www.apperceptual.com/ Peter Turney] | + | * ESL questions available on request from [http://www.apperceptual.com/ Peter Turney] |
| − | * introduced in Turney (2001) as a way of evaluating algorithms for measuring degree of similarity between | + | * introduced in Turney (2001) as a way of evaluating algorithms for measuring degree of similarity between words |
* subsequently used by many other researchers | * subsequently used by many other researchers | ||
| + | * see also: [[Similarity (State of the art)]] | ||
| Line 54: | Line 55: | ||
| Random | | Random | ||
| 25.00% | | 25.00% | ||
| − | | | + | | 14.63-40.34% |
|- | |- | ||
| RES | | RES | ||
| Line 61: | Line 62: | ||
| Hybrid | | Hybrid | ||
| 32.66% | | 32.66% | ||
| − | | | + | | 21.21-48.77% |
|- | |- | ||
| LC | | LC | ||
| Line 68: | Line 69: | ||
| Lexicon-based | | Lexicon-based | ||
| 36.00% | | 36.00% | ||
| − | | | + | | 22.92-50.81% |
|- | |- | ||
| LIN | | LIN | ||
| Line 75: | Line 76: | ||
| Hybrid | | Hybrid | ||
| 36.00% | | 36.00% | ||
| − | | | + | | 22.92-50.81% |
|- | |- | ||
| JC | | JC | ||
| Line 82: | Line 83: | ||
| Hybrid | | Hybrid | ||
| 36.00% | | 36.00% | ||
| − | | | + | | 22.92-50.81% |
|- | |- | ||
| HSO | | HSO | ||
| Line 88: | Line 89: | ||
| Jarmasz and Szpakowicz (2003) | | Jarmasz and Szpakowicz (2003) | ||
| Lexicon-based | | Lexicon-based | ||
| − | | 62.00 | + | | 62.00% |
| − | | | + | | 47.18-75.35% |
|- | |- | ||
| PMI-IR | | PMI-IR | ||
| Line 96: | Line 97: | ||
| Corpus-based | | Corpus-based | ||
| 74.00% | | 74.00% | ||
| − | | | + | | 59.66-85.37% |
| + | |- | ||
| + | | PMI-IR | ||
| + | | Terra and Clarke (2003) | ||
| + | | Terra and Clarke (2003) | ||
| + | | Corpus-based | ||
| + | | 80.00% | ||
| + | | 66.28-89.97% | ||
|- | |- | ||
| JS | | JS | ||
| Line 103: | Line 111: | ||
| Lexicon-based | | Lexicon-based | ||
| 82.00% | | 82.00% | ||
| − | | | + | | 68.56-91.42% |
|- | |- | ||
|} | |} | ||
| + | |||
== Explanation of table == | == Explanation of table == | ||
| Line 111: | Line 120: | ||
* '''Algorithm''' = name of algorithm | * '''Algorithm''' = name of algorithm | ||
* '''Reference for algorithm''' = where to find out more about given algorithm | * '''Reference for algorithm''' = where to find out more about given algorithm | ||
| − | * '''Reference for experiment''' = where to find out more about evaluation of given algorithm with | + | * '''Reference for experiment''' = where to find out more about evaluation of given algorithm with ESL questions |
* '''Type''' = general type of algorithm: corpus-based, lexicon-based, hybrid | * '''Type''' = general type of algorithm: corpus-based, lexicon-based, hybrid | ||
* '''Correct''' = percent of 80 questions that given algorithm answered correctly | * '''Correct''' = percent of 80 questions that given algorithm answered correctly | ||
| − | * '''95% confidence''' = confidence interval calculated using [ | + | * '''95% confidence''' = confidence interval calculated using the [[Statistical calculators|Binomial Exact Test]] |
* table rows sorted in order of increasing percent correct | * table rows sorted in order of increasing percent correct | ||
* several WordNet-based similarity measures are implemented in [http://www.d.umn.edu/~tpederse/ Ted Pedersen]'s [http://www.d.umn.edu/~tpederse/similarity.html WordNet::Similarity] package | * several WordNet-based similarity measures are implemented in [http://www.d.umn.edu/~tpederse/ Ted Pedersen]'s [http://www.d.umn.edu/~tpederse/similarity.html WordNet::Similarity] package | ||
| + | * PMI-IR = Pointwise Mutual Information - Information Retrieval | ||
| + | * Terra and Clarke (2003) call the ESL Synonym Questions "TS1" | ||
| Line 127: | Line 138: | ||
== References == | == References == | ||
| − | Jarmasz, M., and Szpakowicz, S. (2003). [http://www. | + | Hirst, G., and St-Onge, D. (1998). [http://mirror.eacoss.org/documentation/ITLibrary/IRIS/Data/1997/Hirst/Lexical/1997-Hirst-Lexical.pdf Lexical chains as representation of context for the detection and correction of malapropisms]. In C. Fellbaum (ed.), ''WordNet: An Electronic Lexical Database''. Cambridge: MIT Press, 305-332. |
| + | |||
| + | Jarmasz, M., and Szpakowicz, S. (2003). [http://www.csi.uottawa.ca/~szpak/recent_papers/TR-2003-01.pdf Roget’s thesaurus and semantic similarity], ''Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP-03)'', Borovets, Bulgaria, September, pp. 212-219. | ||
| + | |||
| + | Jiang, J.J., and Conrath, D.W. (1997). [http://wortschatz.uni-leipzig.de/~sbordag/aalw05/Referate/03_Assoziationen_BudanitskyResnik/Jiang_Conrath_97.pdf Semantic similarity based on corpus statistics and lexical taxonomy]. ''Proceedings of the International Conference on Research in Computational Linguistics'', Taiwan. | ||
| + | |||
| + | Leacock, C., and Chodorow, M. (1998). [http://books.google.ca/books?id=Rehu8OOzMIMC&lpg=PA265&ots=IpnaLkZUec&lr&pg=PA265#v=onepage&q&f=false Combining local context and WordNet similarity for word sense identification]. In C. Fellbaum (ed.), ''WordNet: An Electronic Lexical Database''. Cambridge: MIT Press, pp. 265-283. | ||
| + | |||
| + | Lin, D. (1998). [http://www.cs.ualberta.ca/~lindek/papers/sim.pdf An information-theoretic definition of similarity]. ''Proceedings of the 15th International Conference on Machine Learning (ICML-98)'', Madison, WI, pp. 296-304. | ||
| + | |||
| + | Resnik, P. (1995). [http://citeseer.ist.psu.edu/resnik95using.html Using information content to evaluate semantic similarity]. ''Proceedings of the 14th International Joint Conference on Artificial Intelligence (IJCAI-95)'', Montreal, pp. 448-453. | ||
Terra, E., and Clarke, C.L.A. (2003). [http://acl.ldc.upenn.edu/N/N03/N03-1032.pdf Frequency estimates for statistical word similarity measures]. ''Proceedings of the Human Language Technology and North American Chapter of Association of Computational Linguistics Conference 2003 (HLT/NAACL 2003)'', pp. 244–251. | Terra, E., and Clarke, C.L.A. (2003). [http://acl.ldc.upenn.edu/N/N03/N03-1032.pdf Frequency estimates for statistical word similarity measures]. ''Proceedings of the Human Language Technology and North American Chapter of Association of Computational Linguistics Conference 2003 (HLT/NAACL 2003)'', pp. 244–251. | ||
Turney, P.D. (2001). [http://arxiv.org/abs/cs.LG/0212033 Mining the Web for synonyms: PMI-IR versus LSA on TOEFL]. ''Proceedings of the Twelfth European Conference on Machine Learning (ECML-2001)'', Freiburg, Germany, pp. 491-502. | Turney, P.D. (2001). [http://arxiv.org/abs/cs.LG/0212033 Mining the Web for synonyms: PMI-IR versus LSA on TOEFL]. ''Proceedings of the Twelfth European Conference on Machine Learning (ECML-2001)'', Freiburg, Germany, pp. 491-502. | ||
| + | |||
| + | |||
| + | [[Category:State of the art]] | ||
| + | [[Category:Similarity]] | ||
Latest revision as of 11:23, 28 June 2015
- ESL = English as a Second Language
- 50 multiple-choice synonym questions; 4 choices per question
- each question includes a sentence, providing context for the question
- ESL questions available on request from Peter Turney
- introduced in Turney (2001) as a way of evaluating algorithms for measuring degree of similarity between words
- subsequently used by many other researchers
- see also: Similarity (State of the art)
Sample question
Stem: "A rusty nail is not as strong as a clean, new one." Choices: (a) corroded (b) black (c) dirty (d) painted Solution: (a) corroded
Table of results
| Algorithm | Reference for algorithm | Reference for experiment | Type | Correct | 95% confidence |
|---|---|---|---|---|---|
| Random | Random guessing | 1 / 4 = 25.00% | Random | 25.00% | 14.63-40.34% |
| RES | Resnik (1995) | Jarmasz and Szpakowicz (2003) | Hybrid | 32.66% | 21.21-48.77% |
| LC | Leacock and Chodrow (1998) | Jarmasz and Szpakowicz (2003) | Lexicon-based | 36.00% | 22.92-50.81% |
| LIN | Lin (1998) | Jarmasz and Szpakowicz (2003) | Hybrid | 36.00% | 22.92-50.81% |
| JC | Jiang and Conrath (1997) | Jarmasz and Szpakowicz (2003) | Hybrid | 36.00% | 22.92-50.81% |
| HSO | Hirst and St.-Onge (1998) | Jarmasz and Szpakowicz (2003) | Lexicon-based | 62.00% | 47.18-75.35% |
| PMI-IR | Turney (2001) | Turney (2001) | Corpus-based | 74.00% | 59.66-85.37% |
| PMI-IR | Terra and Clarke (2003) | Terra and Clarke (2003) | Corpus-based | 80.00% | 66.28-89.97% |
| JS | Jarmasz and Szpakowicz (2003) | Jarmasz and Szpakowicz (2003) | Lexicon-based | 82.00% | 68.56-91.42% |
Explanation of table
- Algorithm = name of algorithm
- Reference for algorithm = where to find out more about given algorithm
- Reference for experiment = where to find out more about evaluation of given algorithm with ESL questions
- Type = general type of algorithm: corpus-based, lexicon-based, hybrid
- Correct = percent of 80 questions that given algorithm answered correctly
- 95% confidence = confidence interval calculated using the Binomial Exact Test
- table rows sorted in order of increasing percent correct
- several WordNet-based similarity measures are implemented in Ted Pedersen's WordNet::Similarity package
- PMI-IR = Pointwise Mutual Information - Information Retrieval
- Terra and Clarke (2003) call the ESL Synonym Questions "TS1"
Caveats
- the performance of a corpus-based algorithm depends on the corpus, so the difference in performance between two corpus-based systems may be due to the different corpora, rather than the different algorithms
- the ESL questions include nouns, verbs, and adjectives, but some of the WordNet-based algorithms were only designed to work with nouns
References
Hirst, G., and St-Onge, D. (1998). Lexical chains as representation of context for the detection and correction of malapropisms. In C. Fellbaum (ed.), WordNet: An Electronic Lexical Database. Cambridge: MIT Press, 305-332.
Jarmasz, M., and Szpakowicz, S. (2003). Roget’s thesaurus and semantic similarity, Proceedings of the International Conference on Recent Advances in Natural Language Processing (RANLP-03), Borovets, Bulgaria, September, pp. 212-219.
Jiang, J.J., and Conrath, D.W. (1997). Semantic similarity based on corpus statistics and lexical taxonomy. Proceedings of the International Conference on Research in Computational Linguistics, Taiwan.
Leacock, C., and Chodorow, M. (1998). Combining local context and WordNet similarity for word sense identification. In C. Fellbaum (ed.), WordNet: An Electronic Lexical Database. Cambridge: MIT Press, pp. 265-283.
Lin, D. (1998). An information-theoretic definition of similarity. Proceedings of the 15th International Conference on Machine Learning (ICML-98), Madison, WI, pp. 296-304.
Resnik, P. (1995). Using information content to evaluate semantic similarity. Proceedings of the 14th International Joint Conference on Artificial Intelligence (IJCAI-95), Montreal, pp. 448-453.
Terra, E., and Clarke, C.L.A. (2003). Frequency estimates for statistical word similarity measures. Proceedings of the Human Language Technology and North American Chapter of Association of Computational Linguistics Conference 2003 (HLT/NAACL 2003), pp. 244–251.
Turney, P.D. (2001). Mining the Web for synonyms: PMI-IR versus LSA on TOEFL. Proceedings of the Twelfth European Conference on Machine Learning (ECML-2001), Freiburg, Germany, pp. 491-502.