Mining semantic networks of bioinformatics e-resources from the literature
2011

Mining Semantic Networks of Bioinformatics E-Resources

Sample size: 2691 publication Evidence: moderate

Author Information

Author(s): Afzal Hammad, Eales James, Stevens Robert, Nenadic Goran

Primary Institution: University of Manchester

Hypothesis

Automated approaches could be used to improve the discovery process by generating semantic networks and clusters of similar bioinformatics resources.

Conclusion

The method can reconstruct interesting functional links between resources, demonstrating the potential of combining literature mining and lexical kernel methods to model relatedness between resource descriptors.

Supporting Evidence

  • The method processed 2,691 full-text bioinformatics articles.
  • 12,452 resources were extracted with associated descriptors.
  • An average of 13.77 descriptors were assigned per resource.

Takeaway

This study shows how we can use computer programs to find and connect different bioinformatics tools and resources by looking at scientific papers.

Methodology

The methodology involves identifying mentions of bioinformatics resources in literature, generating semantic profiles, and linking resources based on similarity metrics.

Limitations

The approach may struggle with resources that appear infrequently and are represented by small descriptor sets.

Digital Object Identifier (DOI)

10.1186/2041-1480-2-S1-S4

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication