{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/distributed-dual-vigilance-fuzzy-adaptive","title":"Distributed dual vigilance fuzzy adaptive resonance theory learns online, retrieves arbitrarily-shaped clusters, and mitigates order dependence","arxiv_id":"1901.00794","date":"2018-11-28","proceeding":null,"authors":["da Silva Leonardo Enzo Brito","Elnabarawy Islam","Wunsch Donald C. II"],"abstract":"This paper presents a novel adaptive resonance theory (ART)-based modular\narchitecture for unsupervised learning, namely the distributed dual vigilance\nfuzzy ART (DDVFA). DDVFA consists of a global ART system whose nodes are local\nfuzzy ART modules. It is equipped with the distinctive features of distributed\nhigher-order activation and match functions, using dual vigilance parameters\nresponsible for cluster similarity and data quantization. Together, these allow\nDDVFA to perform unsupervised modularization, create multi-prototype clustering\nrepresentations, retrieve arbitrarily-shaped clusters, and control its\ncompactness. Another important contribution is the reduction of\norder-dependence, an issue that affects any agglomerative clustering method.\nThis paper demonstrates two approaches for mitigating order-dependence:\npreprocessing using visual assessment of cluster tendency (VAT) or\npostprocessing using a novel Merge ART module. The former is suitable for batch\nprocessing, whereas the latter can be used in online learning. Experimental\nresults in the online learning mode carried out on 30 benchmark data sets show\nthat DDVFA cascaded with Merge ART statistically outperformed the best other\nART-based systems when samples were randomly presented. Conversely, they were\nfound to be statistically equivalent in the offline mode when samples were\npre-processed using VAT. Remarkably, performance comparisons to non-ART-based\nclustering algorithms show that DDVFA (which learns incrementally) was also\nstatistically equivalent to the non-incremental (offline) methods of DBSCAN,\nsingle linkage hierarchical agglomerative clustering (HAC), and k-means, while\nretaining the appealing properties of ART. Links to the source code and data\nare provided. Considering the algorithm's simplicity, online learning\ncapability, and performance, it is an ideal choice for many agglomerative\nclustering applications.","url_abs":"http://arxiv.org/abs/1901.00794v1","url_pdf":"http://arxiv.org/pdf/1901.00794v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"distributed-dual-vigilance-fuzzy-adaptive","repo_url":"https://github.com/ACIL-Group/DDVFA","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"clustering","task_name":"Clustering"},{"task_slug":"quantization","task_name":"Quantization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}