A Novel Hybrid SBM Clustering Method Based on Fuzzy Time Series

• Published in: IEEE access Vol. 11; p. 1
• Main Authors: Zhang, Ren-Long, Liu, Xiao-Hong
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.01.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Analytical models; Clustering; Clustering algorithm; Clustering algorithms; Clustering methods; Data models; Decision making; Fuzzy systems; Fuzzy time series; Machine learning; Mathematical models; Nonparametric frontier; Nonparametric statistics; SBM; Time series; Time series analysis
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2023.3273010

With the development of machine learning algorithm and fuzzy theory, fuzzy clustering algorithm based on time series has received more and more attention. Based on the time series theory and considering the correlation of data attributes, it proposes a novel multivariate fuzzy time series clustering method based on Slacks Based Measure (MFTS-SBM). Compared with traditional fuzzy clustering that it has the ability to deal with fuzziness and uncertainty, the proposed hybrid SBM clustering method employs with input and output items and considers the clustering results and the influencing factors of nonparametric frontier. Thus, it is important for data decision making because decision makers are interested in understanding the changes required to combine input variables in order to classify them into the desired clusters. The simulation experiment results of different samples are given to explain the use and effectiveness of the proposed hybrid SBM clustering method. Therefore, the hybrid method has strong theoretical significance and practical value.