Related Experiment Video
Updated: Apr 8, 2026

A Web Tool for Generating High Quality Machine-readable Biological Pathways
Published on: February 8, 2017
Ambiguity and variability of database and software names in bioinformatics
Geraint Duck1, Aleksandar Kovacevic2, David L Robertson3
1School of Computer Science, The University of Manchester, Oxford Road, Manchester, M13 9PL UK.
Identifying bioinformatics databases and tools in literature is challenging due to name variations. Machine learning shows promise but requires more context for accurate identification of these essential bioinformatics resources.
Area of Science:
- Bioinformatics
- Computational Biology
- Scientific Literature Analysis
Background:
- Lack of systematic tools to identify bioinformatics databases and tools in scientific literature.
- Variability and ambiguity in naming conventions for software and databases pose challenges.
Purpose of the Study:
- To explore variability and ambiguity in database and software mentions.
- To compare dictionary-based and machine learning approaches for identifying these mentions.
Main Methods:
- Development and manual annotation of a corpus (60 full-text documents).
- Evaluation of a baseline dictionary lookup approach.
- Implementation and assessment of a machine learning model for mention identification.
Main Results:
- High variability and ambiguity in database and software mentions were observed.
- Dictionary lookup achieved an F-score of 46%.
- Machine learning achieved F-scores of 63% (strict) and 70% (lenient), with high precision.
Conclusions:
- Identifying database and tool mentions is complex and current repositories are insufficient.
- Machine learning offers improved accuracy but needs enhanced contextual understanding.
- Further research is needed to capture a wider range of mentions effectively.
Related Concept Videos
Genome Annotation and Assembly
Modern Molecular Taxonomy
Applications of Molecular Taxonomy
Evolutionary Relationships through Genome Comparisons
Microbial Classification System
Genomics

