Creating a Lipoprotein Metabolism Ontology for Text Mining
Author Information
Author(s): Alexopoulou Dimitra, Wächter Thomas, Pickersgill Laura, Eyre Cecilia, Schroeder Michael
Primary Institution: Biotechnology Center (BIOTEC), Technische Universität Dresden
Hypothesis
Can automatic term recognition methods effectively support the development of a lipoprotein metabolism ontology for text mining?
Conclusion
Automatic methods can significantly reduce development time and enhance the identification of domain-specific vocabulary.
Supporting Evidence
- The top 50 predicted terms contained up to 89% relevant terms.
- 53% of LMO terms were found in a corpus of 3066 documents.
- Automatic term recognition methods can aid in ontology design.
Takeaway
The researchers built a list of important terms related to lipoprotein metabolism, and found that machines can help find many of these terms quickly.
Methodology
The study involved manual development of an ontology and comparison with terms generated by four automatic term recognition methods.
Limitations
The ontology's coverage was limited by the document set selected, which did not include all relevant terms.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website