Fully unsupervised fault detection and identification based on recursive density estimation and self-evolving cloud-based classifier
- Submitting institution
-
The University of Lancaster
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 155721550
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2014.05.086
- Title of journal
- Neurocomputing
- Article number
- -
- First page
- 289
- Volume
- 150
- Issue
- A
- ISSN
- 0925-2312
- Open access status
- Out of scope for open access requirements
- Month of publication
- February
- Year of publication
- 2015
- 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
-
2
- Research group(s)
-
B - Data Science
- Citation count
- 71
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper breaks with the traditional approach to Fault Detection and Identification which require supervised learning methods or expert knowledge. Instead, we proposed the first ever fully autonomous method for not only detecting faults (based on a drop in density) but, more significantly, of identification of the type of the fault (using clustering). The paper, published in the prestigious Neurocomputing journal, is the basis of a project directly funded by the Silicon Valley based Ford R&I centre ($150k). This project, targeted at autonomous vehicles, uses the same approach to detect unseen scenarios and learn from them automatically.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
