Real-time feature selection technique with concept drift detection using adaptive micro-clusters for data stream mining
- Submitting institution
-
The University of Reading
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 78678
- Type
- D - Journal article
- DOI
-
10.1016/j.knosys.2018.08.007
- Title of journal
- Knowledge-Based Systems
- Article number
- -
- First page
- 205
- Volume
- 161
- Issue
- -
- ISSN
- 0950-7051
- Open access status
- Compliant
- Month of publication
- -
- Year of publication
- 2018
- 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)
-
9 - DSAI
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The paper presents a feature tracking algorithm for tracking the causality of changes of the patterns encoded in a data stream (concept drift) in real-time. The paper is novel as the presented method is the first of its kind for data stream mining. The significance of this paper is that it enables continuous feature selection and thus faster adaptability to concept drift and thus higher accuracy of real-time data mining workflows. The method has wide ranging potential application including in network intrusion detection and financial markets.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -