Open Collections

UBC Research Data

Text mining for neuroanatomy using WhiteText with an updated corpus and a new web application French, Leon; Pavlidis, Paul

Description

We describe the WhiteText project, and its progress towards automatically extracting statements of neuroanatomical connectivity from text. We review progress to date on the three main steps of the project: recognition of brain region mentions, standardization of brain region mentions to neuroanatomical nomenclature, and connectivity statement extraction. We further describe a new version of our manually curated corpus that adds 2,111 connectivity statements from 1,828 additional abstracts. Cross-validation classification within the new corpus replicates results on our original corpus, recalling 67% of connectivity statements at 51% precision. The resulting merged corpus provides 5,208 connectivity statements that can be used to seed species-specific connectivity matrices and to better train automated techniques. Finally, we present a new web application that allows fast interactive browsing of the over 70,000 sentences indexed by the system, as a tool for accessing the data and assisting in further curation. Software and data are freely available at http://www.chibi.ubc.ca/WhiteText/.

Item Metadata

Title	Text mining for neuroanatomy using WhiteText with an updated corpus and a new web application
Creator	French, Leon; Pavlidis, Paul
Contributor	Cuthill, Melissa
Date Created	2015
Date Issued	2019-03-11
Description	We describe the WhiteText project, and its progress towards automatically extracting statements of neuroanatomical connectivity from text. We review progress to date on the three main steps of the project: recognition of brain region mentions, standardization of brain region mentions to neuroanatomical nomenclature, and connectivity statement extraction. We further describe a new version of our manually curated corpus that adds 2,111 connectivity statements from 1,828 additional abstracts. Cross-validation classification within the new corpus replicates results on our original corpus, recalling 67% of connectivity statements at 51% precision. The resulting merged corpus provides 5,208 connectivity statements that can be used to seed species-specific connectivity matrices and to better train automated techniques. Finally, we present a new web application that allows fast interactive browsing of the over 70,000 sentences indexed by the system, as a tool for accessing the data and assisting in further curation. Software and data are freely available at http://www.chibi.ubc.ca/WhiteText/.
Subject	Medicine, Health and Life Sciences; connectome; text mining; natural language processing; information retrieval
Type	Dataset
Notes	http://hdl.handle.net/11272/10570
Date Available	2019-03-11
Provider	University of British Columbia Library
License	CC0 1.0
DOI	10.14288/1.0363908
URI	https://doi.org/10.5683/SP2/4J5NHT
Publisher DOI	https://doi.org/10.5683/SP2/4J5NHT
Rights URI	http://creativecommons.org/publicdomain/zero/1.0
Aggregated Source Repository	Dataverse

Item Media

Item Citations and Data

Licence

CC0 1.0