A fuzzy approach to text classification with two-stage training for ambiguous instances
- Submitting institution
-
Cardiff University / Prifysgol Caerdydd
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 96290658
- Type
- D - Journal article
- DOI
-
10.1109/TCSS.2019.2892037
- Title of journal
- IEEE Transactions on Computational Social Systems
- Article number
- -
- First page
- 227
- Volume
- 6
- Issue
- 2
- ISSN
- 2329-924X
- Open access status
- Compliant
- Month of publication
- March
- Year of publication
- 2019
- URL
-
https://doi.org/10.1109/TCSS.2019.2892037
- 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
- Yes
- Number of additional authors
-
3
- Research group(s)
-
C - Cybersecurity, privacy and human centred computing
- Citation count
- 4
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This interdisciplinary collaboration with the School of Social Sciences proposes advancements to theoretical models of fuzzy methods to improve the classification of subjective text (e.g. hate speech). Validated across a range of hate speech datasets with significant improvements over existing methods, this method is now used by Greater Manchester Police for online hate trend modelling for intervention purposes. The research was supported by a £1.54 million ESRC grant ES/P010695/1 "Centre for Cyberhate Research & Policy: Real-Time Scalable Methods & Infrastructure for Modelling the Spread of Cyberhate on Social Media".
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
