Unveiling dynamics changes: Singular spectrum analysis-based method for detecting concept drift in industrial data streams

Yuyan Zhang, Zhe Liu, Chunjie Yang, Xiaoke Huang, Siwei Lou, Hanwen Zhang, Duojin Yan

Abstract: Highlights•A drift detection framework based on singular spectrum analysis is constructed.•Common dynamics extraction algorithm and noise-robust index are designed.•Drift threshold is set reasonably based on a consistent distribution estimation.•Superiority of the proposed method is demonstrated theoretically and experimentally.