Combining time-series evidence: A complex network model based on a visibility graph and belief entropy

• Published in: Applied intelligence (Dordrecht, Netherlands) Vol. 52; no. 9; pp. 10706 - 10715
• Main Authors: Song, Xingjian, Xiao, Fuyuan
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.07.2022; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Computer Science; Credibility; Electroencephalography; Entropy; Game theory; Machines; Manufacturing; Mechanical Engineering; Probability distribution; Processes; Time series; Visibility; Weighted visibility graph; Evidence theory; Electroencephalogram; Belief entropy; Time series
• ISSN: 0924-669X; 1573-7497
• DOI: 10.1007/s10489-021-02956-5

Combining basic probability assignments (BPAs) with time series is common in real-life cases. Therefore, a new evidence fusion approach based on belief entropy and a visibility graph (BE-VG) is proposed. The approach converts a time-series BPA into a weighted visibility graph (WVG). In addition, some numerical examples are illustrated to illustrate the efficiency and applicability of the proposed method. Finally, to demonstrate the effect of the BE-VG method, the proposed method is applied to electroencephalogram (EEG) dynamic fusion. Experimentally, the results indicate that the BE-VG method is effective and accurate in conducting EEG signal fusion.