Journal
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
Volume 52, Issue 2, Pages 1121-1134Publisher
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TSMC.2020.3012677
Keywords
Labeling; Time series analysis; Heuristic algorithms; Training; Maximum likelihood estimation; Learning systems; Training data; Dynamic unlabeled examples; ensemble classification; multivariate time series (MTS); online learning; partial label (PL) learning
Funding
- National Key Research and Development Program of China [2018AAA0101100, 2017YFB0503702]
- National Natural Science Foundation of China [61876136]
Ask authors/readers for more resources
Traditional classification learning algorithms have limitations in terms of time consumption for large-scale training multivariate time-series (MTS) data, inability to achieve desired classification accuracy with increasing training MTS data, lack of consideration for utilizing unlabeled samples to enhance classifier performance, and inefficiency in updating shapelet-based association rules. Few studies have addressed online classification learning for dynamically added unlabeled examples. To address these issues, this study proposes an online rule-based classifier learning framework that integrates a confidence-based labeling strategy and an online rule-based classifier learning approach. Extensive experiments on ten datasets demonstrate the effectiveness and efficiency of the proposed approach.
Traditional classification learning algorithms have several limitations: 1) they are time consuming for the large-scale training multivariate time-series (MTS) data, and unsuitable for the dynamically added training data; 2) as the number of the training MTS data becomes larger, they could not achieve the desired classification accuracy; 3) most of them do not consider how to make use of the unlabeled samples to enhance the classifier performance; and 4) due to the high dimension of MTS and complex relationship among variables, existing online learning algorithms are not effective to update shapelet-based association rules. Up to now, few work touched online classification learning for dynamically added unlabeled examples. To efficiently address these issues, we propose an online rule-based classifier learning framework on dynamically added unlabeled MTS data (ORCL-U). This framework integrates a confidence-based labeling strategy (CLS) and an online rule-based classifier learning approach (ORBCL). Extensive experiments on ten datasets show the effectiveness and efficiency of our proposed approach.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available