Fei Huang
ABSTRACT
Named entity extraction has been intensively investigated in the past several years. Both statistical approaches and rule-based approaches have achieved satisfactory performance for regular written/spoken language. However when applied to highly informal or ungrammatical languages, e.g., meeting languages, because of the many mismatches in language genre, the performance of existing methods decreases significantly.
In this paper we propose an adaptive method of named entity extraction for meeting understanding. This method combines a statistical model trained from broadcast news data with a cache model built online for ambiguous words, computes their global context name class probability from local context name class probabilities, and integrates name lists information from meeting profiles. Such a fusion of supervised and unsupervised learning has shown improved performance of named entity extraction for meeting applications. When evaluated using manual meeting transcripts, the proposed method demonstrates a 26.07% improvement over the baseline model. Its performance is also comparable to that of the statistical model trained from a small annotated meeting corpus. We are currently applying the proposed method to automatic meeting transcripts.
- D. Appelt, J. Hobbs, D. Israel, and M. Tyson. Fastus: A finite-state processor for information extraction from real world texts. In Proceeding of IJCAI-93, 1993.Google Scholar
- S. Baluja, V. O. Mittal, and R. Sukthankar. Applying machine learning for high performance named-entity extraction. In Pacific Association for Computational Linguistics, 1999.Google Scholar
- D. Bikel, S. Miller, R. Schwarz, and R. Weischedel. Nymble: A high-performance learning name-finder. In Proceedings of Applied Natural Language Processing, pages 194--201, 1997. Google ScholarDigital Library
- N. Chinchor. Overview of muc-7/met-2. In Proceedings of the Seventh Message Understanding Conference(MUC7), http://www.itl.nist.gov/iaui/894.02/related_projects/muc/proceedings/muc_7_proceedings/overview.html, 1998.Google Scholar
- Y. Gotoh and S. Renals. Information extraction from broadcast news. In Philosophical Transactions of the Royal Society of London, A 358, pages 1295--1310, 2000.Google Scholar
- R. Grishman and B. Sundheim. Design of the muc-6 evaluation. In Proceedings of MUC-6, 1995. Google ScholarDigital Library
- R. Kuhn and R. D. Mori. A cache-based natural language model for speech recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-12(6):570--583, 1990. Google ScholarDigital Library
- N. Chinchor, P. Robinson, and E. Brown. The hub-4 named entity task definition, version 4.8. In Proceedings of DARPA Broadcast News Workshop http://www.nist.gov/speech/hub4_98, 1999.Google Scholar
- L. Rabiner. A tutorial on hidden markov models and selected applications in speech recognition. Proceedings of the IEEE, 77(2):257--285, Feburary, 1989.Google ScholarCross Ref
- P. Robinson, E. Brown, J. Burger, N. Chinchor, A. Douthat, L. Ferro, and L. Hirschman. Overview: Information extraction from broadcast news. In Proceedings of DARPA Broadcast News Workshop, pages 27--30, 1999.Google Scholar
- M. Stevenson and R. Gaizauskas. Using corpus-driven name lists for name entity recognition. In Proceedings of 6th Applied Natural Language Processing and 1st North American Chapter of the Association for Computational Linguistics, 2000. Google ScholarDigital Library
- A. Viterbi. Error bound for convolutional codes and asymptotically optimum decoding algorithm. IEEE Transaction on Information Theory, 13:260--269, 1967.Google ScholarDigital Library
- K. Zechner. Automatic summarization of spoken dialogs in unrestricted domains. In Ph.D Thesis, Language Technology Institute, Carnegie Mellon University, 2001.Google Scholar
Recommendations
Comparison of Methods to Annotate Named Entity Corpora
The authors compared two methods for annotating a corpus for the named entity (NE) recognition task using non-expert annotators: (i) revising the results of an existing NE recognizer and (ii) manually annotating the NEs completely. The annotation time, ...
Named entity extraction and disambiguation: the missing link
ESAIR '13: Proceedings of the sixth international workshop on Exploiting semantic annotations in information retrievalNamed entity extraction (NEE) and disambiguation (NED) are two areas of research that are well covered in literature. Typical fields addressing these topics are information retrieval, natural language processing, and semantic web. Although these topics ...
Automatic gazette creation for named entity recognition and application to resume processing
COMPUTE '12: Proceedings of the 5th ACM COMPUTE Conference: Intelligent & scalable system technologiesNamed entities are important content-carrying units within documents. Consequently named entity recognition (NER) is an important part of information extraction. One fast and accurate approach to NER uses a list or gazette consisting of known instances. ...
Comments