TY - JOUR
T1 - Online semi-supervised multi-channel time series classifier based on growing neural gas
AU - Nooralishahi, Parham
AU - Seera, Manjeevan
AU - Loo, Chu Kiong
N1 - Funding Information:
Acknowledgments The project is funded under the University of Malaya Grand Challenge Grant (GC003A-14HTM).
Funding Information:
The project is funded under the University of Malaya Grand Challenge Grant (GC003A-14HTM).
Publisher Copyright:
© 2016, The Natural Computing Applications Forum.
Copyright:
Copyright 2017 Elsevier B.V., All rights reserved.
PY - 2017/11
Y1 - 2017/11
N2 - Challenges in time series classification has attracted attention in the past decade. Although large amounts of labeled data are assumed to be available, in reality, labeled data might be scarce to find in many domains. In this paper, we propose an online semi-supervised multi-channel classifier for time series based on growing neural gas (GNG) learning scheme. The method is able to handle multi-channel time series with variation in dimensions and it introduces a label prediction strategy to minimize misclassification. It measures the similarity of input instance and learned templates using weighted multi-channel dynamic time warping technique and learns the topology of input data space specified for each class using the GNG learning algorithm. Comprehensive evaluation is conducted using various datasets, such as gesture recognition, human activity recognition, and human daily-life activity recognition. Experimental results demonstrate good classification results, with indication that the proposed approach requires only a handful of labeled instances to construct an accurate classification model.
AB - Challenges in time series classification has attracted attention in the past decade. Although large amounts of labeled data are assumed to be available, in reality, labeled data might be scarce to find in many domains. In this paper, we propose an online semi-supervised multi-channel classifier for time series based on growing neural gas (GNG) learning scheme. The method is able to handle multi-channel time series with variation in dimensions and it introduces a label prediction strategy to minimize misclassification. It measures the similarity of input instance and learned templates using weighted multi-channel dynamic time warping technique and learns the topology of input data space specified for each class using the GNG learning algorithm. Comprehensive evaluation is conducted using various datasets, such as gesture recognition, human activity recognition, and human daily-life activity recognition. Experimental results demonstrate good classification results, with indication that the proposed approach requires only a handful of labeled instances to construct an accurate classification model.
KW - Clustering algorithm
KW - Dynamic time warping
KW - Growing neural gas
KW - Label prediction
KW - Multi-channel time series
KW - Topology formation
UR - http://www.scopus.com/inward/record.url?scp=84960335909&partnerID=8YFLogxK
U2 - 10.1007/s00521-016-2247-2
DO - 10.1007/s00521-016-2247-2
M3 - Article
AN - SCOPUS:84960335909
SN - 0941-0643
VL - 28
SP - 3491
EP - 3505
JO - Neural Computing and Applications
JF - Neural Computing and Applications
IS - 11
ER -