Identifying biological concepts from a protein-related corpus with a probabilistic topic model
2006

Identifying Biological Concepts from Protein-Related Texts

publication Evidence: moderate

Author Information

Author(s): Zheng Bin, McLean David C Jr, Lu Xinghua

Primary Institution: Medical University of South Carolina

Hypothesis

The goal of this research is to identify the major biological topics/concepts from a corpus of protein-related MEDLINE© titles and abstracts by applying a probabilistic topic model.

Conclusion

The major and recurring biological concepts within a collection of MEDLINE documents can be extracted by the LDA model.

Supporting Evidence

  • The LDA model was able to extract 300 major topics from the corpus.
  • Most identified topics were semantically coherent and represented biological objects or concepts.
  • The identified topics can be used to index text effectively.

Takeaway

This study shows how we can find important topics about proteins in scientific papers using a special computer model.

Methodology

The latent Dirichlet allocation (LDA) model was applied to extract 300 major topics from a corpus of MEDLINE titles and abstracts.

Limitations

The LDA model requires a specified number of topics, which may not reflect the true complexity of the data.

Digital Object Identifier (DOI)

10.1186/1471-2105-7-58

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication