Attributed Subspace Clustering
- Submitting institution
-
University of Greenwich
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 30485
- Type
- E - Conference contribution
- DOI
-
10.24963/ijcai.2019/516
- Title of conference / published proceedings
- Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence
- First page
- 3719
- Volume
- 0
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- -
- Year of publication
- 2019
- URL
-
-
- 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
-
5
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Existing subspace clustering (SC) methods are merely able to produce a single clustering solution, which faces limitations because data can often be interpreted from different aspects and grouped into multiple clusters. Therefore, we propose an innovative model called attributed subspace clustering (ASC), which simultaneously learns multiple self-representations from latent representations derived from original data. In comparison with the existing SC approaches, ASC not only improves clustering accuracy by the integrated representation but also achieves multiple clustering solutions, which is beyond what current approaches can offer. The acceptance rate at IJCAI 2019 was 17.9%.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -