A genetic graph-based approach for partitional clustering
- Submitting institution
-
Middlesex University
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 352
- Type
- D - Journal article
- DOI
-
10.1142/S0129065714300083
- Title of journal
- International Journal of Neural Systems
- Article number
- 1430008
- First page
- -
- Volume
- 24
- Issue
- 3
- ISSN
- 0129-0657
- Open access status
- Out of scope for open access requirements
- Month of publication
- January
- Year of publication
- 2014
- URL
-
http://eprints.mdx.ac.uk/28818/
- 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)
-
-
- Citation count
- 63
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work solves the stability problem of spectral clustering, a well-known clustering algorithm with the ability to identify manifolds. The evolutionary graph-based clustering algorithm presented here obtains better manifold identification than spectral clustering and several other state of the art algorithms on real and benchmark datasets. This is significant because it was also the first graph-based genetic optimization algorithm for manifold clustering. It has been extended to six different algorithms in the literature, and it has been successfully applied to behavioral modelling, sport prediction and image segmentation.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -