Abstract:The ensemble of fuzzy clustering can address the problems presented in the base clustering, such as fluctuations in results due to random initialization and performance degradation due to outliers. However, the performance of fuzzy clustering ensembles is still hampered by some challenges that include misaligned membership matrices, loss of information in the co-similarity matrix, large storage space, unstable ensemble results due to an additional re-clustering, the need for original data information for assistance, etc. To address these issues, we propose a parameter-free robust ensemble framework for fuzzy clustering. After obtaining the set of membership matrices, we cascade these membership matrices and mine the latent spectral matrix of the raw data. Benefiting from this step, we obtain global features of the dataset without knowing the specific data. Then, our framework uses transition matrices to solve the alignment problem, avoiding the storage of large-scale matrices. Most importantly, we introduce a robust weighted mechanism in the optimization model, where each base clustering is adaptively adjusted and the effect of outliers is suppressed by a robust function. In addition, the model yields the results as a membership matrix, which produces the exact partition results directly without any subsequent clustering operations. Finally, since our model is a parameter-free model, the setting of hyper-parameters is avoided and the applicability of the model is improved as well. The effective algorithm of the optimization model is derived and its time complexity and convergence are analyzed. The results of competitive experiments on benchmark data show that the proposed ensemble framework is effective compared to state-of-the-art methods.

An Ensemble of Adaptive Neuro-Fuzzy Kohonen Networks for Online Data Stream Fuzzy Clustering

Adaptive Fuzzy Clustering of Multivariate Short Time Series with Unevenly Distributed Observations Based on Matrix Neuro-Fuzzy Self-organizing Network

A Density-Based Competitive Data Stream Clustering Network with Self-Adaptive Distance Metric

ESA-Stream: Efficient Self-Adaptive Online Data Stream Clustering

Density Based Self Organizing Incremental Neural Network For Data Stream Clustering

Online Fuzzy Cognitive Map Learning

An ensemble-based online learning algorithm for streaming data

Online Sparse Representation Clustering for Evolving Data Streams

HClustream: A Novel Approach for Clustering Evolving Heterogeneous Data Stream

Distributed Data Stream Clustering: A Fast EM-based Approach

Online Clustering of Evolution Data Stream Based on Affinity Propagation Clustering

Adaptive Approach to Fuzzy Clustering

Incremental Interval Type-2 Fuzzy Clustering of Data Streams using Single Pass Method

FGCH: a Fast and Grid Based Clustering Algorithm for Hybrid Data Stream

A New Parallel Adaptive Clustering and its Application to Streaming Data

Research on the seawater quality monitoring technology based on the fuzzy clustering

Dynamic Micro-cluster-Based Streaming Data Clustering Method for Anomaly Detection.

Evolving Restricted Boltzmann Machine-Kohonen Network for Online Clustering

Online Stream Clustering Using Density and Affinity Propagation Algorithm

Topology Learning-Based Fuzzy Random Neural Networks for Streaming Data Regression

Parameter-Free Robust Ensemble Framework of Fuzzy Clustering