Dynamic summarization of bibliographic-based data
2011

Dynamic Summarization of Bibliographic Data

Sample size: 667 publication Evidence: high

Author Information

Author(s): Workman T Elizabeth, Hurdle John F

Primary Institution: University of Utah

Hypothesis

The MEDLINE output based on a user-generated PubMed query that is constrained to the desired topic and point-of-view will generate SemRep output that likely contains a semantic profile representative of the same topic and point-of-view focus.

Conclusion

The new Combo algorithm outperformed a conventional summarization schema in a genetic database curation task.

Supporting Evidence

  • Combo identified 74 genetic entities implicated in bladder cancer development, outperforming traditional methods.
  • The Combo algorithm achieved 61% recall and 81% precision in identifying relevant genetic entities.
  • The conventional schema only identified 10 genetic entities, highlighting the effectiveness of the new algorithm.

Takeaway

This study created a smart tool that helps find important information about bladder cancer from a lot of medical articles quickly.

Methodology

The study developed a new algorithm called Combo that combines three statistical metrics to identify salient data in bibliographic text, evaluated against a conventional schema.

Limitations

The study's performance cannot be quantified in other applications until similar research is conducted.

Digital Object Identifier (DOI)

10.1186/1472-6947-11-6

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication