Low-Rank Tensor Regularized Fuzzy Clustering for Multiview Data

Wei, Huiqin; Chen, Long; Ruan, Keyu; Li, Lingxi

doi:10.1109/tfuzz.2020.2988841

Cited by 21 publications

(2 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The NMI, ACC, and ARI values are in the range of [0, 1], [0, 1], and [−1, 1], respectively. [29], divergence-based locally weighted ensemble clustering with dictionary learning and L2,1-Norm (DL-WECDL) [30], Parameter-Free Robust Ensemble Framework of Fuzzy Clustering (PFREFF) [31], parameter-free consensus embedding learning for multiview graph-based clustering (PF-CEL) [10], and robust and fuzzy ensemble framework via spectral learning for random projection-based fuzzy-cmeans clustering (FESRPF) [6].…”

Section: B Evaluation Metricsmentioning

confidence: 99%

Fuzzy Divergence Weighted Ensemble Clustering With Spectral Learning Based on Random Projections for Big Data

Lahmar,

Zaier,

Yahia

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In many real-world applications, data are described by high-dimensional feature spaces, posing new challenges for current ensemble clustering methods. The goal is to combine sets of base clusters to enhance clustering accuracy, but this makes them susceptible to low quality. However, the reliability of present ensemble clustering in high-dimensional data still needs improvement. In this context, we propose a new fuzzy divergence-weighted ensemble clustering based on random projection and spectral learning. Firstly, random projection (RP) is used to create various dimensional data and find membership matrices via fuzzy c-means (FCM). Secondly, fuzzy partitions of random projections are ranked using entropy-based local weighting along with Kullback-Leibler (KL) divergence to detect any uncertainty. Then it used to evaluate the weight of each cluster. Finally, we create regularized graphs from these membership matrices and use spectral matrices to estimate the affinity matrices of these graphs using fuzzy KL divergence anchor graphs. Subsequently, obtaining the final clustering results is considered as an optimization problem, and the ensemble clustering results are obtained. The experimental results on high-dimensional data demonstrate the efficiency of our method compared to state-of-the-art methods.

show abstract

Section: B Evaluation Metricsmentioning

confidence: 99%

Fuzzy Divergence Weighted Ensemble Clustering With Spectral Learning Based on Random Projections for Big Data

Lahmar,

Zaier,

Yahia

et al. 2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The abnormal interference data of laser seeker guidance can be processed by detection and classification methods. At present, the methods used to eliminate outlier data based on classification can be divided as the neural network methods [9] [10], discriminant analysis methods [11], clustering methods [12], support vector machine methods [13], and so on. Yuen et al [14] adopted a probability method for outlier detection and quantified the outlier probability of data points, considering not only the optimal values of parameters and residuals, but also the uncertainty of data.…”

Section: Introductionmentioning

confidence: 99%

Abnormal Data Detection of Guidance Angle Based On SMP-SVDD For Seeker

Liang

Zhang

Cui

et al. 2021

Preprint

View full text Add to dashboard Cite

The accuracy of the pitch angle deviation directly affects the guidance accuracy of the laser seeker. During the guidance process, the abnormal pitch angle deviation data will be produced when the seeker is affected by interference sources. In this paper, aiming to detect abnormal data in seeker pitch angle deviation data, a method based on Smooth Multi-Kernel Polarization Support Vector Data Description (SMP-SVDD) is proposed to detect abnormal data in guidance angle data. On the one hand, the polarization value is used to determine the weight of the multi-kernel combination coefficient to obtain the multi-kernel polarization function, and the particle swarm optimization is used to find the optimal kernel, which improves the detection accuracy. On the other hand, the constrained quadratic programming problem is smooth and differentiable, and the conjugate gradient method can be applied to reduce the complexity of problem solving. Through simulation experiments, it is verified that the SMP-SVDD method has higher detection accuracy and faster calculation speed compared with different detection methods in different guidance stages.

show abstract