A dynamic linear model for heteroscedastic LDA under class imbalance
- Submitting institution
-
Coventry University
- Unit of assessment
- 12 - Engineering
- Output identifier
- 21407663
- Type
- D - Journal article
- DOI
-
10.1016/j.neucom.2018.07.090
- Title of journal
- Neurocomputing
- Article number
- -
- First page
- 65
- Volume
- 343
- Issue
- -
- ISSN
- 0925-2312
- Open access status
- Exception within 3 months of publication
- Month of publication
- February
- 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
-
3
- Research group(s)
-
-
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- Funded by National Engineering Laboratory (Brian.Millington@tuv-sud.co.uk), the research developed a theoretically founded variant of linear discriminant classifier (LDA) that extends LDA to unequal variances without requiring quadratic terms. The classifier is robust to class imbalance, as shown by the validation with 8 real-world data sets. It is fast, computationally light, and suitable for diverse applications from remote health monitoring to machine health monitoring, where an “event/fault” state is not as probable as the “normal” state of the system. Work to utilize the method for industrial flow meter diagnosis is ongoing with NEL.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
