Large-scale directional relationship extraction and resolution
2008

Extracting Directional Relationships in Biomedical Literature

Sample size: 17145 publication 10 minutes Evidence: moderate

Author Information

Author(s): Giles Cory B, Wren Jonathan D

Primary Institution: Oklahoma Medical Research Foundation

Hypothesis

Can automated methods accurately extract directional relationships from large-scale biomedical literature?

Conclusion

The study demonstrates that while thesaurus-based directional relation extraction can achieve reasonable accuracy, it is susceptible to false positives due to noun modifiers.

Supporting Evidence

  • The SVM classifier achieved 82% precision and 94.8% recall on gold standard corpora.
  • Directional relationships were often extracted with ambiguity.
  • Contextual factors influenced the interpretation of relationships.

Takeaway

The researchers created a computer program to find out how different biological things affect each other by reading a lot of scientific papers. They found that sometimes the results can be confusing.

Methodology

The study used a support vector machine (SVM) to classify dependency paths for extracting relationships from biomedical text.

Potential Biases

There is a risk of bias due to the reliance on noun modifiers and the context in which relationships are reported.

Limitations

The method is prone to false positives and may struggle with context-dependent relationships.

Statistical Information

P-Value

0.05

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-9-S9-S11

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication