Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2008 Nov 19;9 Suppl 11(Suppl 11):S4.
doi: 10.1186/1471-2105-9-S11-S4.

Cascaded classifiers for confidence-based chemical named entity recognition

Peter Corbett  1 , Ann Copestake
Affiliations

Cascaded classifiers for confidence-based chemical named entity recognition

Peter Corbett et al. BMC Bioinformatics. .

Abstract

Background: Chemical named entities represent an important facet of biomedical text.

Results: We have developed a system to use character-based n-grams, Maximum Entropy Markov Models and rescoring to recognise chemical names and other such entities, and to make confidence estimates for the extracted entities. An adjustable threshold allows the system to be tuned to high precision or high recall. At a threshold set for balanced precision and recall, we were able to extract named entities at an F score of 80.7% from chemistry papers and 83.2% from PubMed abstracts. Furthermore, we were able to achieve 57.6% and 60.3% recall at 95% precision, and 58.9% and 49.1% precision at 90% recall.

Conclusion: These results show that chemical named entities can be extracted with good performance, and that the properties of the extraction can be tuned to suit the demands of the task.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Evaluation on chemistry papers.
Figure 2
Figure 2
Evaluation on PubMed abstracts.
Figure 3
Figure 3
Evaluation on chemistry papers, showing effects of disallowing overlapping entities.
Figure 4
Figure 4
Evaluation on chemistry papers, showing performance on different named entity classes.
Figure 5
Figure 5
Evaluation on PubMed abstracts, showing performance on different named entity classes.
See this image and copyright information in PMC

References

    1. Reyle U. Understanding chemical terminology. Terminology. 2006;12:111–126.
    1. Corbett P, Batchelor C, Teufel S. Annotation of Chemical Named Entities. Proceedings of BioNLP 2007: Biological, translational, and clinical language processing (ACL 2007 workshop) 2007. pp. 57–64.
    1. Wilbur WJ, Hazard G, Divita G, Mork J, Aronson A, Browne A. Analysis of Biomedical Text for Chemical Names: A Comparison of Three Methods. Proceedings of the Annual Fall Symposium of the American Medical Informatics Association (AMIA 99) 1999. pp. 176–180. - PMC - PubMed
    1. Vasserman A. Identifying Chemical Names in Biomedical Text: an Investigation of Substring Co-occurrence Based Approaches. Proceedings of the Student Research Workshop at HLT-NAACL. 2004. pp. 7–12.
    1. Townsend J, Copestake A, Murray-Rust P, Teufel S, Waudby C. Language Technology for Processing Chemistry Publications. Proceedings of the UK e-Science All Hands Meeting. 2005. pp. 247–253.

Publication types

LinkOut - more resources