TY - JOUR
T1 - Meta-cognitive recurrent recursive kernel OS-ELM for concept drift handling
AU - Liu, Zongying
AU - Loo, Chu Kiong
AU - Seera, Manjeevan
N1 - Funding Information:
The authors express their gratitude to The Twin Industrial Park (project RP025B-15HNE ), Thailand Research Fund (grant agreement TRG5680090 ), and ONRG grant (Project No.: ONRG - NICOP - N62909-18-1-2086 ) from Office of Naval Research Global, UK , in supporting this research.
Publisher Copyright:
© 2018 Elsevier B.V.
Copyright:
Copyright 2019 Elsevier B.V., All rights reserved.
PY - 2019/2
Y1 - 2019/2
N2 - In this paper, a Meta-cognitive Recurrent Recursive Kernel Online Sequential Extreme Learning Machine with Drift Detector Mechanism (meta-RRKOS-ELM-DDM) is proposed. It combines the strengths of Recurrent Kernel Online Sequential Extreme Learning Machine with a new modified Drift Detector Mechanism (DDM) and Approximate Linear Dependency Kernel Filter (ALD) in solving concept drift problems and reducing complex computations in the learning. The recursive kernel method successfully replaces the normal kernel method in Recurrent Kernel Online Sequential Extreme Learning Machine with DDM (RKOS-ELM-DDM) and generates a fixed reservoir with optimized information in enhancing the forecasting performance. Meta-cognitive learning strategy decides when the incoming data needs to be updated, retrained, or discarded during learning and automatically finding ALD threshold that reduces the learning time of prediction model. In the experiment, six synthetic and three real-world time series datasets are used to evaluate the ability of recursive kernel method, the performance of concept drift detectors, and meta-cognitive learning strategy in time series prediction. Experimental results indicate the meta-RRKOS-ELM with DDM has superior prediction ability in the different predicting horizons as compared with other algorithms.
AB - In this paper, a Meta-cognitive Recurrent Recursive Kernel Online Sequential Extreme Learning Machine with Drift Detector Mechanism (meta-RRKOS-ELM-DDM) is proposed. It combines the strengths of Recurrent Kernel Online Sequential Extreme Learning Machine with a new modified Drift Detector Mechanism (DDM) and Approximate Linear Dependency Kernel Filter (ALD) in solving concept drift problems and reducing complex computations in the learning. The recursive kernel method successfully replaces the normal kernel method in Recurrent Kernel Online Sequential Extreme Learning Machine with DDM (RKOS-ELM-DDM) and generates a fixed reservoir with optimized information in enhancing the forecasting performance. Meta-cognitive learning strategy decides when the incoming data needs to be updated, retrained, or discarded during learning and automatically finding ALD threshold that reduces the learning time of prediction model. In the experiment, six synthetic and three real-world time series datasets are used to evaluate the ability of recursive kernel method, the performance of concept drift detectors, and meta-cognitive learning strategy in time series prediction. Experimental results indicate the meta-RRKOS-ELM with DDM has superior prediction ability in the different predicting horizons as compared with other algorithms.
KW - Concept drift
KW - Kernel adaptive filter
KW - Kernel Online Sequential Extreme Learning Machine
KW - Meta-cognitive learning
KW - Time series prediction
UR - http://www.scopus.com/inward/record.url?scp=85057830138&partnerID=8YFLogxK
U2 - 10.1016/j.asoc.2018.11.006
DO - 10.1016/j.asoc.2018.11.006
M3 - Article
AN - SCOPUS:85057830138
VL - 75
SP - 494
EP - 507
JO - Applied Soft Computing
JF - Applied Soft Computing
SN - 1568-4946
ER -