Multi-component nonnegative matrix factorization
- Submitting institution
-
University of Greenwich
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 30481
- Type
- E - Conference contribution
- DOI
-
10.24963/ijcai.2017/407
- Title of conference / published proceedings
- Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence
- First page
- 2922
- Volume
- 0
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- -
- Year of publication
- 2017
- 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
- Nonnegative matrix factorization (NMF) is a core technique to find data representations to support clustering and classification. Real-world data are usually complex and contain independent yet interlinked dimensions. Failing to capture the complementary information among them may lead to an inaccurate representation. To overcome this limitation, we propose a novel multi-component nonnegative matrix factorization (MCNMF). It not only achieves greater accuracy than the state-of-the-art in terms of clustering and classification, but also interprets data semantically, which extends the representational capability of NMFs. The acceptance rate at IJCAI 2017 was 26%.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
