A Text Mining Pipeline Using Active and Deep Learning Aimed at Curating Information in Computational Neuroscience
- Submitting institution
-
Manchester Metropolitan University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2358
- Type
- D - Journal article
- DOI
-
10.1007/s12021-018-9404-y
- Title of journal
- Neuroinformatics
- Article number
- -
- First page
- 391
- Volume
- 17
- Issue
- 3
- ISSN
- 1539-2791
- Open access status
- Compliant
- Month of publication
- November
- Year of publication
- 2018
- URL
-
https://e-space.mmu.ac.uk/622400/
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
6
- Research group(s)
-
A - Data Science
- Citation count
- 3
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work combines a state-of-the-art deep learning pipeline with a novel corpus of Neuroscience entities. The corpus contains 1000 annotated sentences using active learning, and almost 4000 annotated entities. The research shows marked improvements over previous work considered to be state-of-the-art and extends these improvements beyond previously studies categories of neuroscience entities. This research was undertaken with neuroscience domain experts from the Blue Brain Project at EPFL, Lausanne (Renaud Richardet, Christian O’Reilly). Using the framework proposed in this paper, researchers from the neuroinformatics community have analysed the neuroscience literature to detect and categorise structured information.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -