Enhancing access to the Bibliome: the TREC 2004 Genomics Track
2006

Improving Information Retrieval in Genomics

Sample size: 27 publication Evidence: moderate

Author Information

Author(s): Williams R. Hersh, Ravi Teja Bhupatiraju, Laura Ross, Phoebe Roberts, Aaron M. Cohen, Dale F. Kraemer

Primary Institution: Oregon Health & Science University

Hypothesis

The TREC Genomics Track aims to enhance information retrieval in genomics through the creation of test collections.

Conclusion

The TREC Genomics Track provides diverse approaches to ad hoc retrieval, leading to varied efficacy in information retrieval systems.

Supporting Evidence

  • A total of 27 research groups submitted 47 different runs.
  • The best MAP obtained by any run was 0.4075.
  • Techniques that expanded queries with gene name lists had the best efficacy.
  • Many runs performed more poorly than a simple baseline run.

Takeaway

This study is about making it easier for scientists to find information in a huge pile of research papers about genes and biology.

Methodology

The study involved an ad hoc retrieval task using a subset of the MEDLINE database, with relevance judgments made by biologists.

Potential Biases

Potential bias in relevance judgments due to the limited number of judges and their backgrounds.

Limitations

The evaluation using test collections may not reflect real-world user interactions and relevance judgments can be inconsistent.

Participant Demographics

Participants included 27 research groups from various institutions.

Statistical Information

P-Value

0.4075

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1747-5333-1-3

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication