NP Chunking (State of the art)
- Performance measure: F = 2 * Precision * Recall / (Recall + Precision)
- Precision: percentage of NPs found by the algorithm that are correct
- Recall: percentage of NPs defined in the corpus that were found by the chunking program
- Training data: sections 15-18 of Wall Street Journal corpus (Ramshaw and Marcus)
- Testing data: section 20 of Wall Street Journal corpus (Ramshaw and Marcus)
- original data of the NP chunking experiments by Lance Ramshaw and Mitch Marcus
- data contains one word per line and each line contains six fields of which only the first three fields are relevant: the word, the part-of-speech tag assigned by the Brill tagger, and the correct IOB tag
Table of results
|System name||Short description||Main publications||Software||Reports (F)|
|KM00||B-I-O tagging using SVM classifiers with polynomial kernel||Kudo and Matsumoto (2000), CONLL||sbobet (but models are not provided)||93.79%|
|KM01||learning as in KM00, but voting between different representations||Kudo and Matsumoto (2001), NAACL||No||94.22%|
|SP03||Second order conditional random fields||Fei Sha and Fernando Pereira (2003), HLT/NAACL||No||94.3%|
|SS05||specialized HMM + voting between different representations||Shen and Sarkar (2005)||No||95.23%|
|M05||Second order conditional random fields + multi-label classification||Ryan McDonald, KOby Crammer and Fernando Pereira (2005), HLT/EMNLP||No||94.29%|
|V06||Conditional random fields + Stochastic Meta Decent (SMD)||S. V. N. Vishwanathan, Nicol N. Schraudolph, Mark Schmidt, and Kevin Murphy (2006), ICML||No||93.6%|
|S08||Second order latent-dynamic conditional random fields + an improved inference method based on A* search||Xu Sun, Louis-Philippe Morency, Daisuke Okanohara and Jun'ichi Tsujii (2008), COLING||HCRF Library||94.34%|
|C00||Chunks from the Charniak Parser||Hollingshead, Fisher and Roark (2005), Charniak (2000)||?||94.20%|
E. Charniak (2000). A Maximum-Entropy inspired parser, NAACL 2000
K. Hollingshead, S. Fisher and B. Roark (2005). Comparing and combining finite-state and context-free parsers. HLT/EMNLP 2005.
T. Kudo and Y. Matsumoto (2000). Use of support vector learning for chunk identification. Proceedings of the 4th Conference on CoNLL-2000 and LLL-2000, pages 142-144, Lisbon, Portugal.
T. Kudo and Y. Matsumoto (2001). Chunking with support vector machines. Proceedings of NAACL-2001.
F. Sha and F. Pereira (2003). Shallow Parsing with Conditional Random Fields. Proceedings of HLT-NAACL 2003, pages 213-220. Edmonton, Canada.
H. Shen and A. Sarkar (2005). Voting between multiple data representations for text chunking. Proceedings of the Eighteenth Meeting of the Canadian Society for Computational Intelligence, Canadian AI 2005.
R. McDonald, K. Crammer and F. Pereira (2005). Flexible Text Segmentation with Structured Multilabel Classification. Human Language Technologies and Empirical Methods in Natural Language Processing (HLT-EMNLP), 2005
S. V. N. Vishwanathan, N. Schraudolph, M. Schmidt, and K. Murphy. Accelerated Training Conditional Random Fields with Stochastic Gradient Methods. In Proc. Intl. Conf. Machine Learning, pp. 969 – 976, ACM Press, New York, NY, USA, 2006.
X. Sun, L.P. Morency, D. OKanohara and J. Tsujii (2008). Modeling Latent-Dynamic in Shallow Parsing: A Latent Conditional Model with Improved Inference. Proceedings of The 22nd International Conference on Computational Linguistics (COLING 2008). Pages 841-848. Manchester, UK.