Extracting Protein Annotations from Biomedical Articles
Author Information
Author(s): Krallinger Martin, Padron Maria, Valencia Alfonso
Primary Institution: National Center of Biotechnology, CNB-CSIC
Hypothesis
Can a sentence sliding window approach effectively extract protein annotations from biomedical literature?
Conclusion
The sentence sliding window approach efficiently extracts protein annotations, achieving a high number of correct predictions.
Supporting Evidence
- The method achieved 28.8% correct predictions in the BioCreative contest.
- The approach outperformed other techniques in extracting protein annotations.
- The study highlights the need for efficient automatic filtering in biomedical literature.
Takeaway
This study shows a way to automatically find important information about proteins in scientific articles, making it easier for researchers.
Methodology
The study used a sentence sliding window technique to score text passages for protein annotations based on their context.
Potential Biases
Potential biases in the evaluation process due to the subjective nature of annotation assessments.
Limitations
The dataset used for training was noisy and not fully representative, which may have affected the results.
Statistical Information
P-Value
p<0.05
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website