Open AccessSome properties of the classification algorithms using ensemble kernels
Author(s) -
Nikita Odinokikh,
Vladimir Berikov
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1715/1/012011
Subject(s) - computer science , kernel (algebra) , cluster (spacecraft) , algorithm , ensemble learning , noise (video) , point (geometry) , statistical classification , quality (philosophy) , artificial intelligence , pattern recognition (psychology) , data mining , mathematics , philosophy , geometry , epistemology , combinatorics , image (mathematics) , programming language
In this paper, we study the properties of the KCCE algorithm (Kernel-based Classification with Cluster Ensemble) proposed in our previous publication. Different strategies for choosing the ensemble weights are investigated from the point of view of their influence on the quality and speed of the algorithm. We perform numerical experiments with the algorithm on real datasets from the UCI repository under conditions of artificially added noise at different levels.