Semantically linking and browsing PubMed abstracts with gene ontology
2008

Linking PubMed Abstracts to Gene Ontology Using Semantics

Sample size: 491 publication Evidence: moderate

Author Information

Author(s): Vanteru Bhanu C, Shaik Jahangheer S, Yeasin Mohammed

Primary Institution: University of Memphis

Hypothesis

Can a semantic-sensitive search engine improve the retrieval of relevant PubMed abstracts linked to Gene Ontology?

Conclusion

The proposed SEGOPubmed technique outperformed traditional keyword-based methods in associating relevant abstracts to Gene Ontology terms.

Supporting Evidence

  • SEGOPubmed incorporates semantics into the ontology-based searching of PubMed.
  • The analysis using well-referenced keywords shows that the proposed technique outperformed string comparison methods.
  • SEGOPubmed extracted abstracts where keywords appeared in combination with other terms, which traditional methods missed.

Takeaway

This study created a smarter way to find research papers by understanding the meaning of words, not just matching them exactly.

Methodology

The study used Latent Semantic Analysis to semantically link PubMed abstracts to Gene Ontology terms and compared its performance with existing methods.

Limitations

The study is limited to a few well-referenced keywords and does not address the concept of polysemy.

Digital Object Identifier (DOI)

10.1186/1471-2164-9-S1-S10

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication