Use artificial neural network to align biological ontologies
2008

Aligning Biological Ontologies Using Artificial Neural Networks

Sample size: 120 publication Evidence: moderate

Author Information

Author(s): Huang Jingshan, Dang Jiangbo, Huhns Michael N, Zheng W Jim

Primary Institution: Medical University of South Carolina

Hypothesis

Three weights for semantic aspects learned from a subset of concepts are representative of all concepts in the same ontology.

Conclusion

The experiment results verify that the proposed method significantly improves the automation of biological ontology alignment.

Supporting Evidence

  • The method achieved a Precision of 0.9 and Recall of 0.85.
  • Three weights for semantic aspects were learned and validated through experiments.
  • Two real-world biological ontologies were used for testing the proposed method.

Takeaway

This study shows how we can use computers to help match different biological terms, making it easier for scientists to understand and share information.

Methodology

An artificial neural network approach was used to learn and adjust weights for semantic aspects in biological ontology alignment.

Potential Biases

Potential bias in the selection of training examples and the reliance on domain experts for validation.

Limitations

The study primarily focuses on finding equivalent concept pairs and does not address other mapping tasks.

Statistical Information

P-Value

0.9 and 0.85 for Precision and Recall respectively

Statistical Significance

p<0.05

Digital Object Identifier (DOI)

10.1186/1471-2164-9-S2-S16

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication