PubMed related articles: a probabilistic topic-based model for content similarity
2007

A New Model for Finding Related Articles in PubMed

Sample size: 4584 publication Evidence: moderate

Author Information

Author(s): Lin Jimmy, Wilbur W John

Primary Institution: University of Maryland, College Park, Maryland, USA

Hypothesis

The pmra model will improve the precision of related article searches in PubMed compared to the bm25 model.

Conclusion

The pmra model provides a small but statistically significant improvement in precision for related article searches in PubMed.

Supporting Evidence

  • The pmra model showed a statistically significant improvement over the bm25 model in terms of precision.
  • Approximately 19% of user sessions involved at least one click on a related article.
  • The study utilized a test collection from the TREC 2005 genomics track to evaluate the models.

Takeaway

This study created a new way to find articles related to a specific topic in PubMed, which helps researchers find more useful information.

Methodology

The pmra model was tested against the bm25 model using a test collection from the TREC 2005 genomics track, focusing on precision at five retrieved documents.

Limitations

The study's results may not generalize beyond the biomedical domain due to the reliance on MeSH terms for parameter estimation.

Statistical Information

P-Value

0.022

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2105-8-423

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication