Output details
11 - Computer Science and Informatics
University of Westminster
Article title
Support vector machine learning with an evolutionary engine
Type
D - Journal article
Title of journal
Journal of the Operational Research Society
Article number
-
Volume number
60
Issue number
8
First page of article
1116
ISSN of journal
0160-5682
Year of publication
2009
Number of additional authors
4
Additional information
<15> Originality: This paper presents a novel hybrid approach encompassing the geometrical consideration of learning within support vector machines while it considers the estimation for the coefficients of the decision surface through the direct search capabilities of evolutionary algorithms.
Significance: The proposed approach resolves the complexity of the optimizer opens the black-box of support vector training and breaks the limits of the canonical solving component.
Rigour: Computational results show that our approach is as efficient as the standard canonical approach in terms of running time yet offers a greater flexibility.
Interdisciplinary
-
Cross-referral requested
-
Research group
None
Citation count
9
Proposed double-weighted
No
Double-weighted statement
-
Reserve for a double-weighted output
No
Non-English
No
English abstract
-
