Infinite Latent Feature Selection: A Probabilistic Latent Graph-Based Ranking Approach
- Submitting institution
-
University of Glasgow
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-01083
- Type
- E - Conference contribution
- DOI
-
10.1109/ICCV.2017.156
- Title of conference / published proceedings
- IEEE International Conference on Computer Vision (ICCV 2017)
- First page
- 1407
- Volume
- -
- Issue
- -
- ISSN
- 2380-7504
- Open access status
- Compliant
- Month of publication
- December
- Year of publication
- 2017
- URL
-
http://eprints.gla.ac.uk/149366/
- 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
-
3
- Research group(s)
-
-
- Citation count
- 58
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- ORIGINALITY: First feature selection approach using topic models to quantify how beneficial every feature is in a pattern recognition task. RIGOUR: Method was systematically compared to 11 state- of-the-art methods using 10 benchmarks (always with positive results). SIGNIFICANCE: implementation is available on MATLAB File Exchange (https://uk.mathworks.com/matlabcentral/fileexchange/56937-feature-selection-library#feedbacks). Downloaded 17,864 times since 2016, it attracted an average rating of 4.82 out of 5.00 (60+ ratings) and was recognised as an outstanding contribution by MATLAB in 2017. It was invited for presentation to the MathWorks Research Summit, an “invitation-only event […] that redefines the frontiers of research” (https://uk.mathworks.com/videos/series/mathworks-research-summit.html).
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -