Extract interaction detection methods from the biological literature
2009

Extracting Interaction Detection Methods from Biological Literature

Sample size: 5319 publication Evidence: high

Author Information

Author(s): Wang Hongning, Huang Minlie, Zhu Xiaoyan

Primary Institution: Tsinghua University

Hypothesis

Can a generative topic model effectively extract interaction detection methods from biological literature?

Conclusion

The CMW model successfully captures the correlations between detection methods and related words, outperforming previous methods.

Supporting Evidence

  • The CMW model outperformed the best result reported in the BioCreative II challenge evaluation.
  • The model effectively captures the in-depth correlations between detection methods and related words.
  • Using a large corpus of 5319 documents, the model demonstrated competitive performance against traditional classifiers.

Takeaway

This study created a smart model that helps find different ways scientists describe how proteins interact, making it easier to understand their research.

Methodology

The study developed a generative topic model called the Correlated Method-Word model to extract detection methods from a large corpus of documents.

Limitations

The model may struggle with the diversity of method descriptions and relies on the quality of the input data.

Digital Object Identifier (DOI)

10.1186/1471-2105-10-S1-S55

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication