Overview of BioCreAtIvE: Critical Assessment of Information Extraction for Biology
Author Information
Author(s): Lynette Hirschman, Alexander Yeh, Christian Blaschke, Alfonso Valencia
Primary Institution: The MITRE Corporation
Hypothesis
The goal of the first BioCreAtIvE challenge was to provide a set of common evaluation tasks to assess the state of the art for text mining applied to biological problems.
Conclusion
The first BioCreAtIvE assessment achieved a high level of international participation and provided state-of-the-art performance results for gene name finding and normalization, while highlighting limitations in functional annotation tasks.
Supporting Evidence
- The assessment provided state-of-the-art performance results for gene name finding and normalization.
- The best systems achieved a balanced 80% precision/recall or better.
- The results for functional annotation were significantly lower, demonstrating current limitations.
Takeaway
BioCreAtIvE is a project that helps scientists figure out how well computers can read and understand biology papers, especially when it comes to finding names of genes and proteins.
Methodology
The assessment involved two main tasks: extracting gene or protein names from text and identifying text passages that support Gene Ontology annotations.
Limitations
The results for the advanced task of functional annotation were significantly lower, indicating limitations in current text-mining approaches.
Participant Demographics
27 groups from 10 countries participated in the assessment.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website