Connecting Targets to Tweets: Semantic Attention-based Model for Target-Specific stance Detection
- Submitting institution
-
University of Durham
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 116731
- Type
- E - Conference contribution
- DOI
-
10.1007/978-3-319-68783-4_2
- Title of conference / published proceedings
- Web Information Systems Engineering – WISE 2017, 18th International Conference
- First page
- 18
- Volume
- -
- Issue
- -
- ISSN
- 03029743
- Open access status
- Technical exception
- Month of publication
- -
- Year of publication
- 2017
- URL
-
https://doi.org/10.1007/978-3-319-68783-4_2
- 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
-
2
- Research group(s)
-
A - Innovative Computing
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work, supported by the Alan Turing Institute, combines for the first time different information features, from target and source, into the first deep learning solution to outperform traditional methods in stance detection. Importantly, this represents a new method of embedding semantics into the ‘black-box’ neural network. The paper provides a systematic theoretical explanation of the approach, detailed network architecture descriptions, and clear practical applications, without relying on engineered features and thus enhancing generalisability. This work led to an Alan Turing Fellowship, collaboration with HSBC, and a position with Amazon for the PhD student.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -