SuRF: Identification of Interesting Data Regions with Surrogate Models
- Submitting institution
-
University of Glasgow
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 11-09954
- Type
- E - Conference contribution
- DOI
-
10.1109/ICDE48307.2020.00118
- Title of conference / published proceedings
- 36th IEEE International Conference on Data Engineering (IEEE ICDE)
- First page
- 1321
- Volume
- -
- Issue
- -
- ISSN
- 2375-026X
- Open access status
- Compliant
- Month of publication
- May
- Year of publication
- 2020
- URL
-
http://eprints.gla.ac.uk/209812/
- 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
- 0
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- ORIGINALITY: The paper proposes a novel evolutionary machine learning algorithm for advanced data mining/exploratory tasks and was published at the premier conference on Data Engineering systems. SIGNIFICANCE: This work has laid the foundation of the query-driven evolutionary ML by significantly minimizing the data access required for mining data regions in decentralized data systems forming the base of modern data mining. RIGOUR: The paper contains analytical formulation of novel evolutionary multi-modal optimization models that effectively and efficiently mine data regions of interest supporting exploratory tasks evaluated using a combination of simulation and testbed experiments in real commercial data systems.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -