MKEM: a Multi-level Knowledge Emergence Model for mining undiscovered public knowledge
2010

MKEM: A Model for Discovering Hidden Knowledge in Biomedical Literature

Sample size: 5000 publication Evidence: moderate

Author Information

Author(s): Ijaz Ali Z, Min Song, Doheon Lee

Primary Institution: KAIST, South Korea

Hypothesis

Can a Multi-level Knowledge Emergence Model (MKEM) effectively uncover undiscovered public knowledge in biomedical literature?

Conclusion

The MKEM model is effective in discovering hidden relationships in biomedical texts.

Supporting Evidence

  • The system extracted 410 relationships from 5000 abstracts.
  • Precision of the system was 75%, indicating a good level of accuracy.
  • 24 new hypotheses were generated from the extracted data.

Takeaway

This study created a computer program that helps find new connections in medical research by looking at lots of articles at once.

Methodology

The study used Natural Language Processing techniques and a model called SEPDB to extract relationships from 5000 biomedical abstracts.

Limitations

The performance evaluation lacked a gold standard for comparison.

Digital Object Identifier (DOI)

10.1186/1471-2105-11-S2-S3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication