Back to Search
Start Over
MeSHLabeler: improving the accuracy of large-scale MeSH indexing by integrating diverse evidence
- Source :
- Bioinformatics
- Publication Year :
- 2015
- Publisher :
- Oxford University Press (OUP), 2015.
-
Abstract
- Motivation: Medical Subject Headings (MeSHs) are used by National Library of Medicine (NLM) to index almost all citations in MEDLINE, which greatly facilitates the applications of biomedical information retrieval and text mining. To reduce the time and financial cost of manual annotation, NLM has developed a software package, Medical Text Indexer (MTI), for assisting MeSH annotation, which uses k-nearest neighbors (KNN), pattern matching and indexing rules. Other types of information, such as prediction by MeSH classifiers (trained separately), can also be used for automatic MeSH annotation. However, existing methods cannot effectively integrate multiple evidence for MeSH annotation. Methods: We propose a novel framework, MeSHLabeler, to integrate multiple evidence for accurate MeSH annotation by using ‘learning to rank’. Evidence includes numerous predictions from MeSH classifiers, KNN, pattern matching, MTI and the correlation between different MeSH terms, etc. Each MeSH classifier is trained independently, and thus prediction scores from different classifiers are incomparable. To address this issue, we have developed an effective score normalization procedure to improve the prediction accuracy. Results: MeSHLabeler won the first place in Task 2A of 2014 BioASQ challenge, achieving the Micro F-measure of 0.6248 for 9,040 citations provided by the BioASQ challenge. Note that this accuracy is around 9.15% higher than 0.5724, obtained by MTI. Availability and implementation: The software is available upon request. Contact: zhusf@fudan.edu.cn
- Subjects :
- Statistics and Probability
Abstracting and Indexing
Computer science
MEDLINE
Ismb/Eccb 2015 Proceedings Papers Committee July 10 to July 14, 2015, Dublin, Ireland
Machine learning
computer.software_genre
Biochemistry
Medical Subject Headings
Annotation
Text mining
Data Mining
Pattern matching
Molecular Biology
Data
business.industry
Search engine indexing
Reproducibility of Results
Computer Science Applications
Computational Mathematics
Computational Theory and Mathematics
Learning to rank
Artificial intelligence
Data mining
business
computer
Classifier (UML)
Algorithms
Software
Subjects
Details
- ISSN :
- 13674811 and 13674803
- Volume :
- 31
- Database :
- OpenAIRE
- Journal :
- Bioinformatics
- Accession number :
- edsair.doi.dedup.....d25625dbc252c56ee339aa0abb59da59