Disambiguation of biomedical text using diverse sources of information
2008

Disambiguating Biomedical Text with Multiple Information Sources

Sample size: 100 publication Evidence: moderate

Author Information

Author(s): Mark Stevenson, Yikun Guo, Robert Gaizauskas, David Martinez

Primary Institution: Department of Computer Science, University of Sheffield

Hypothesis

Can combining various sources of information improve the disambiguation of biomedical terms?

Conclusion

Using a combination of linguistic features and MeSH terms significantly improves the disambiguation of biomedical terms.

Supporting Evidence

  • The best performance was achieved using a combination of linguistic features and MeSH terms.
  • Performance exceeded previously published results for systems evaluated using the same data set.
  • Statistical tests confirmed the significance of the improvements observed.

Takeaway

This study shows that using different types of information helps computers understand medical texts better, especially when they use special terms called MeSH.

Methodology

The study compared various information sources for word sense disambiguation in biomedical texts using a standard test set.

Limitations

The study did not explore the hierarchical structure of MeSH terms, which could provide additional insights.

Statistical Information

P-Value

p<0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-9-S11-S7

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication