Adaptive Resonance Theory-based Topological Clustering with a Divisive Hierarchical Structure Capable of Continual Learning

• Published in: IEEE access Vol. 10; p. 1
• Main Authors: Masuyama, Naoki, Amako, Narito, Yamada, Yuna, Nojima, Yusuke, Ishibuchi, Hisao
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptive Resonance Theory; Algorithms; Big Data; Cluster analysis; Clustering; Clustering algorithms; Continual Learning; Data mining; Data points; Hierarchical Clustering; Information retrieval; Kernel; Machine learning; Partitioning algorithms; Resonance; Similarity; Structural hierarchy; Subspace constraints; Topological Clustering; Topology; Training data
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2022.3186479

Adaptive Resonance Theory (ART) is considered as an effective approach for realizing continual learning thanks to its ability to handle the plasticity-stability dilemma. In general, however, the clustering performance of ART-based algorithms strongly depends on the specification of a similarity threshold, i.e., a vigilance parameter, which is data-dependent and specified by hand. This paper proposes an ART-based topological clustering algorithm with a mechanism that automatically estimates a similarity threshold from the distribution of data points. In addition, for improving information extraction performance, a divisive hierarchical clustering algorithm capable of continual learning is proposed by introducing a hierarchical structure to the proposed algorithm. Experimental results demonstrate that the proposed algorithm has high clustering performance comparable with recently-proposed state-of-the-art hierarchical clustering algorithms.