Incremental Cluster Validity Index-Guided Online Learning for Performance and Robustness to Presentation Order

被引:4
作者
Brito da Silva, Leonardo Enzo [1 ]
Rayapati, Nagasharath [1 ]
Wunsch, Donald C., II [2 ]
机构
[1] Guise Inc, Rolla, MO 65401 USA
[2] Missouri Univ Sci & Technol, Dept Elect & Comp Engn, Rolla, MO 65401 USA
关键词
Adaptive resonance theory predictive mapping (ARTMAP); clustering; data streams; incremental cluster validity index (iCVI); online learning; semi-supervised learning; FUZZY ARTMAP; CLASSIFICATION; ARCHITECTURE; EXTENSIONS;
D O I
10.1109/TNNLS.2022.3212345
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In streaming data applications, the incoming samples are processed and discarded, and therefore, intelligent decision-making is crucial for the performance of lifelong learning systems. In addition, the order in which the samples arrive may heavily affect the performance of incremental learners. The recently introduced incremental cluster validity indices (iCVIs) provide valuable aid in addressing such class of problems. Their primary use case has been cluster quality monitoring; nonetheless, they have been recently integrated in a streaming clustering method. In this context, the work presented, here, introduces the first adaptive resonance theory (ART)-based model that uses iCVIs for unsupervised and semi-supervised online learning. Moreover, it shows how to use iCVIs to regulate ART vigilance via an iCVI-based match tracking mechanism. The model achieves improved accuracy and robustness to ordering effects by integrating an online iCVI module as module B of a topological ART predictive mapping (TopoARTMAP)-thereby being named iCVI-TopoARTMAP-and using iCVI-driven post-processing heuristics at the end of each learning step. The online iCVI module provides assignments of input samples to clusters at each iteration in accordance to any of the several iCVIs. The iCVI-TopoARTMAP maintains useful properties shared by the ART predictive mapping (ARTMAP) models, such as stability, immunity to catastrophic forgetting, and the many-to-one mapping capability via the map field module. The performance and robustness to the presentation order of iCVI-TopoARTMAP were evaluated via experiments with synthetic and real-world datasets.
引用
收藏
页码:6686 / 6700
页数:15
相关论文
共 72 条
[1]   An extensive comparative study of cluster validity indices [J].
Arbelaitz, Olatz ;
Gurrutxaga, Ibai ;
Muguerza, Javier ;
Perez, Jesus M. ;
Perona, Inigo .
PATTERN RECOGNITION, 2013, 46 (01) :243-256
[2]  
BARTFAI G, 1994, 1994 IEEE INTERNATIONAL CONFERENCE ON NEURAL NETWORKS, VOL 1-7, P940, DOI 10.1109/ICNN.1994.374307
[3]   Streaming Data Analysis: Clustering or Classification? [J].
Bezdek, James C. ;
Keller, James M. .
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS, 2021, 51 (01) :91-102
[4]   VAT: A tool for visual assessment of (cluster) tendency [J].
Bezdek, JC ;
Hathaway, RJ .
PROCEEDING OF THE 2002 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-3, 2002, :2225-2230
[5]   An evolving approach to data streams clustering based on typicality and eccentricity data analytics [J].
Bezerra, Clauber Gomes ;
Jales Costa, Bruno Sielly ;
Guedes, Luiz Affonso ;
Angelov, Plamen Parvanov .
INFORMATION SCIENCES, 2020, 518 :13-28
[6]   Incremental Cluster Validity Indices for Online Learning of Hard Partitions: Extensions and Comparative Study [J].
Brito Da Silva, Leonardo Enzo ;
Melton, Niklas Max ;
Wunsch, Donald C. .
IEEE ACCESS, 2020, 8 :22025-22047
[7]   A survey of adaptive resonance theory neural network models for engineering applications [J].
Brito da Silva, Leonardo Enzo ;
Elnabarawy, Islam ;
Wunsch, Donald C., II .
NEURAL NETWORKS, 2019, 120 :167-203
[8]   Distributed dual vigilance fuzzy adaptive resonance theory learns online, retrieves arbitrarily-shaped clusters, and mitigates order dependence [J].
Brito da Silva, Leonardo Enzo ;
Elnabarawy, Islam ;
Wunsch, Donald C., II .
NEURAL NETWORKS, 2020, 121 :208-228
[9]   Dual vigilance fuzzy adaptive resonance theory [J].
Brito da Silva, Leonardo Enzo ;
Elnabarawy, Islam ;
Wunsch, Donald C., II .
NEURAL NETWORKS, 2019, 109 :1-5
[10]  
da Silva LEB, 2017, 2017 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (SSCI), P1358