One-step multi-view spectral clustering with cluster label correlation graph

Hajjar, S. El; Dornaika, Fadi; Abdallah, Fahed

doi:10.1016/j.ins.2022.01.017

Cited by 34 publications

(2 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sally El Hajjar and Fadi Dornaika's paper [15] outlines an innovative method for doing graph-based multiview clustering in one step. Two significant advancements are included, both of which set it apart from other graph-based approaches that cluster in a single step.…”

Section: International Journal On Recent and Innovation Trends Inmentioning

confidence: 99%

An Analytical Performance Evaluation on Multiview Clustering Approaches

Mankar,

Kamalapur

2023

IJRITCC

View full text Add to dashboard Cite

The concept of machine learning encompasses a wide variety of different approaches, one of which is called clustering. The data points are grouped together in this approach to the problem. Using a clustering method, it is feasible, given a collection of data points, to classify each data point as belonging to a specific group. This can be done if the algorithm is given the collection of data points. In theory, data points that constitute the same group ought to have attributes and characteristics that are equivalent to one another, however data points that belong to other groups ought to have properties and characteristics that are very different from one another. The generation of multiview data is made possible by recent developments in information collecting technologies. The data were collected from à variety of sources and were analysed using a variety of perspectives. The data in question are what are known as multiview data. On a single view, the conventional clustering algorithms are applied. In spite of this, real-world data are complicated and can be clustered in a variety of different ways, depending on how the data are interpreted. In practise, the real-world data are messy. In recent years, Multiview Clustering, often known as MVC, has garnered an increasing amount of attention due to its goal of utilising complimentary and consensus information derived from different points of view. On the other hand, the vast majority of the systems that are currently available only enable the single-clustering scenario, whereby only makes utilization of a single cluster to split the data. This is the case since there is only one cluster accessible. In light of this, it is absolutely necessary to carry out investigation on the multiview data format. The study work is centred on multiview clustering and how well it performs compared to these other strategies.

show abstract

Section: International Journal On Recent and Innovation Trends Inmentioning

confidence: 99%

An Analytical Performance Evaluation on Multiview Clustering Approaches

Mankar,

Kamalapur

2023

IJRITCC

View full text Add to dashboard Cite

show abstract

“…Leveraging advancements in signal processing and data analysis techniques, spectral clustering has emerged as a promising tool for addressing these challenges [19]. This powerful clustering method, rooted in graph theory, excels at identifying clusters within complex data structures [20][21][22]. By mapping data points into a lower dimensional space, spectral clustering simplifies and streamlines clustering tasks.…”

Section: Introductionmentioning

confidence: 99%

Ice Cover Prediction for Transmission Lines Based on Feature Extraction and an Improved Transformer Scheme

Ke,

Sun,

Zhao

et al. 2024

Electronics

View full text Add to dashboard Cite

Frequent and severe icing on transmission lines poses a serious threat to the stability and safe operation of the power system. Meteorological data, inherently stochastic and uncertain, requires effective preprocessing and feature extraction to ensure accurate and efficient prediction of transmission line icing thickness. We address this challenge by leveraging the meteorological features of icing phenomena and propose a novel feature preprocessing method that integrates Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and spectral clustering. This method effectively preprocesses raw time series data, extracts key features, and constructs a similarity matrix and feature vector. The resulting feature vector serves as a new data representation, facilitating cluster analysis to isolate meteorological and icing-related features specific to transmission lines. Subsequently, we introduce an enhanced Transformer model for predicting transmission line icing thickness. The proposed model leverages the extracted meteorological and icing features by independently embedding variable tokens for each input feature. This approach improves the model’s prediction accuracy under multiple feature inputs, leading to more effective learning. The experimental results demonstrate that the performance of the proposed prediction algorithm is better than the three baseline algorithms (hybrid CEEMDAN and LSTM (CEEMDAN-LSTM), hybrid CEEMDAN, spectral clustering, and LSTM (CEEMDAN-SP-LSTM), and hybrid CEEMDAN, spectral clustering, and Transformer (CEEMDAN-SP-Transformer)) under multiple feature inputs and different parameter settings.

show abstract