Supervised Named Entity Recognition Algorithm
- AKA: Supervised NER Method.
- It can be composed of:
- It can range from being a Fully-Supervised Named Entity Recognition Algorithm to being a Semi-Supervised Named Entity Recognition Algorithm.
- It can make use of an NER Predictor Feature.
- See: Supervised WSD Algorithm.
- (Ratinov & Roth, 2009) ⇒ Lev Ratinov, and Dan Roth. (2009). “Design Challenges and Misconceptions in Named Entity Recognition.” In: Proceedings of the Thirteenth Conference on Computational Natural Language Learning.
- QUOTE: We have presented a simple model for NER that uses expressive features to achieve new state of the art performance on the Named Entity recognition task. We explored four fundamental design decisions: text chunks representation, inference algorithm, using non-local features and external knowledge. We showed that BILOU encoding scheme significantly outperforms BIO and that, surprisingly, a conditional model that does not take into account interactions at the output level performs comparably to beam-search or Viterbi, while being considerably more efficient computationally. We analyzed several approaches for modeling non-local dependencies and found that none of them clearly outperforms the others across several datasets. Our experiments corroborate recently published results indicating that word class models learned on unlabeled text can be an alternative to the traditional semi-supervised learning paradigm. NER proves to be a knowledge-intensive task, and it was reassuring to observe that knowledge-driven techniques adapt well across several domains. We observed consistent performance gains across several domains, most interestingly in Webpages, where the named entities had less context and were different in nature from the named entities in the training set. Our system significantly outperforms the current state of the art and is available to download under a research license.
- (Nadeau & Sekine, 2007) ⇒ David Nadeau, and Satoshi Sekine. (2007). “A Survey of Named Entity Recognition and Classification.” In: Lingvisticae Investigationes, 30(1).
- (Tasi et al., 2006) ⇒ Tzong-han Tsai, Wen-Chi Chou, Shih-Hung Wu, Ting-Yi Sung, Jieh Hsiang, and Wen-Lian Hsu. (2006). “Integrating Linguistic Knowledge into a Conditional Random Field Framework to Identify Biomedical Named Entities.” In: Expert Systems with Applications: An International Journal, 30(1). doi:10.1016/j.eswa.2005.09.072
- In the NER problem, we regard each word in a sentence as a token. Each token is associated with a tag that indicates the category of the NE and the location of the token within the NE, for example, B_c, I_c where [math]c[/math] is a category. These two tags denote respectively the beginning token and the following token of an NE in category c. In addition, we use the tag $O$ to indicate that a token is not part of an NE. The NER problem can then be phrased as the problem of assigning one of 2n + 1 tags to each token, where n is the number of NE categories. In the JNLPBA 2004 task, there are 5 named entity categories and 11 tags. For example, one way to tag the phrase IL-2 gene expression, CD28, and NFkappa B in a paper is “B-DNA, I-DNA, O, O, B-protein, O, O, B-protein, I-protein”.
- (Minkov et al., 2005) ⇒ Einat Minkov, Richard C. Wang, and William W. Cohen. (2005). “Extracting personal names from email: applying named entity recognition to informal text.” In: Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing. doi:10.3115/1220575.1220631
- (McDonald et al., 2004) ⇒ Ryan T. McDonald, R. Scott Winters, Mark Mandel, Yang Jin, Peter S. White and Fernando Pereira. (2004). “An entity tagger for recognizing acquired genomic variations in cancer literature." Bioinformatics 2004 20(17):3249-3251; doi:10.1093/bioinformatics/bth350
- (McCallum & Li, 2003) ⇒ Andrew McCallum, and Wei Li. (2003). “Early Results for Named Entity Recognition with Conditional Random Fields, Feature Induction and Web-Enhanced Lexicons.” In: Proceedings of Seventh Conference on Natural Language Learning (CoNLL 2003).
- (FEOAL, 2002) ⇒ K Franzén, G Eriksson, F Olsson, L Asker, P Lidén, J. Coster. (2002). “Protein names and how to find them." Elsevier. International Journal of Medical Informatics, Volume 67, Issue 1 - 3, Pages 49 - 61
- Investigates NER of proteins
- (Bikel et al., 1997) ⇒ Daniel Bikel, Scott Miller, Richard Schwartz, and Ralph Weischedel. (1997). “Nymble: a High-performance Learning Name-finder.” In: Proceedings of Fifth Applied Natural Language Processing Conference (ANLC 1997). doi:10.3115/974557.974586
- One of the earlier examples where learning was competitive with manually coded systems.