Cluster Information Research Articles

Advancing Data Privacy: A Novel K-Anonymity Algorithm with Dissimilarity Tree-Based Clustering and Minimal Information Loss

Anonymization serves as a crucial privacy protection technique employed across various technology domains, including cloud storage, machine learning, data mining and big data to safeguard sensitive information from unauthorized third-party access. As the significance and volume of data grow exponentially, comprehensive data protection against all threats is of utmost importance. The main objective of this paper is to provide a brief summary of techniques for data anonymization and differential privacy.A new k-anonymity method, which deviates from conventional k-anonymity approaches, is proposed by us to address privacy protection concerns. Our paper presents a new algorithm designed to achieve k-anonymity through more efficient clustering. The processing of data by most clustering algorithms requires substantial computation. However, by identifying initial centers that align with the data structure, a superior cluster arrangement can be obtained.Our study presents a Dissimilarity Tree-based strategy for selecting optimal starting centroids and generating more accurate clusters with reduced computing time and Normalised Certainty Penalty (NCP). This method also has the added benefit of reducing the Normalised Certainty Penalty (NCP). When compared to other methods, the graphical performance analysis shows that this one reduces the amount of overall information lost in the dataset being anonymized by around 20% on average. In addition, the method that we have designed is capable of properly handling both numerical and category characteristics.

Ensemble clustering via fusing global and local structure information

Ensemble clustering is aimed at obtaining a robust consensus result from a set of weak base clusterings. Most existing methods rely on a co-association (CA) matrix that describes the frequency at which pairs of samples are clustered into the same class and exhibits a symmetrical property. However, these methods typically focus on either the global or local structure of the CA matrix and do not consider using both types of information simultaneously to improve ensemble clustering performance. To address this issue, we propose a novel scheme to Fuse both the global Structure and the local structure information for Ensemble Clustering, namely FSEC, in this paper. Specifically, FSEC integrates both global structural information and local structural information into a learning framework by self-expressive and CA matrix self-enhancement models respectively. Moreover, FSEC embeds a Hadamard product fusion term to maximize the commonalities between global structure and local structure. The objective function optimization problem is solved by using the alternating direction method of multipliers (ADMM). Experimental results demonstrate that the proposed FSEC outperforms many state-of-the-art methods of ensemble clustering.

Laplacian-based Cluster-Contractive t-SNE for High-Dimensional Data Visualization

Dimensionality reduction techniques aim at representing high-dimensional data in low-dimensional spaces to extract hidden and useful information or facilitate visual understanding and interpretation of the data. However, few of them take into consideration the potential cluster information contained implicitly in the high-dimensional data. In this article, we propose Lap tSNE, a new graph-layout nonlinear dimensionality reduction method based on t-SNE, one of the best techniques for visualizing high-dimensional data as 2D scatter plots. Specifically, Lap tSNE leverages the eigenvalue information of the graph Laplacian to shrink the potential clusters in the low-dimensional embedding when learning to preserve the local and global structure from high-dimensional space to low-dimensional space. It is nontrivial to solve the proposed model because the eigenvalues of normalized symmetric Laplacian are functions of the decision variable. We provide a majorization-minimization algorithm with convergence guarantee to solve the optimization problem of Lap tSNE and show how to calculate the gradient analytically, which may be of broad interest when considering optimization with Laplacian-composited objective. We evaluate our method by a formal comparison with state-of-the-art methods on seven benchmark datasets, both visually and via established quantitative measurements. The results demonstrate the superiority of our method over baselines such as t-SNE and UMAP. We also provide out-of-sample extension, large-scale extension, and mini-batch extension for our Lap tSNE to facilitate dimensionality reduction in various scenarios.

Cluster-aware channel estimation with deep learning method in deep-water acoustic communications.

In underwater acoustic (UWA) communications, channels often exhibit a clustered-sparse structure, wherein most of the channel impulse responses are near zero, and only a small number of nonzero taps assemble to form clusters. Several algorithms have used the time-domain sparse characteristic of UWA channels to reduce the complexity of channel estimation and improve the accuracy. Employing the clustered structure to enhance channel estimation performance provides another promising research direction. In this work, a deep learning-based channel estimation method for UWA orthogonal frequency division multiplexing (OFDM) systems is proposed that leverages the clustered structure information. First, a cluster detection model based on convolutional neural networks is introduced to detect the cluster of UWA channels. This method outperforms the traditional Page test algorithm with better accuracy and robustness, particularly in low signal-to-noise ratio conditions. Based on the cluster detection model, a cluster-aware distributed compressed sensing channel estimation method is proposed, which reduces the noise-induced errors by exploiting the joint sparsity between adjacent OFDM symbols and limiting the search space of channel delay spread. Numerical simulation and sea trial results are provided to illustrate the superior performance of the proposed approach in comparison with existing sparse UWA channel estimation methods.

Consensus-based clustering and data aggregation in decentralized network of multi-agent systems.

Multi-agent systems are promising for applications in various fields. However, they require optimization algorithms that can handle large number of agents and heterogeneously connected networks in clustered environments. Planning algorithms performed in the decentralized communication model and clustered environment require precise knowledge about cluster information by compensating noise from other clusters. This article proposes a decentralized data aggregation algorithm using consensus method to perform COUNT and SUM aggregation in a clustered environment. The proposed algorithm introduces a trust value to perform accurate aggregation on cluster level. The correction parameter is used to adjust the accuracy of the solution and the computation time. The proposed algorithm is evaluated in simulations with large and sparse networks and small bandwidth. The results show that the proposed algorithm can achieve convergence on the aggregated data with reasonable accuracy and convergence time. In the future, the proposed tools will be useful for developing a robust decentralized task assignment algorithm in a heterogeneous multi-agent multi-task environment.

Cluster-aware attentive convolutional recurrent network for multivariate time-series forecasting

Multivariate time-series (MTS) forecasting plays a crucial role in various real-world applications, but the complex dependencies between time-series variables (i.e., inter-series dependencies) make this task extremely challenging. While most existing studies focus on modeling intra-series (temporal) dependencies by capturing long- and short-term patterns, they fail to explore and exploit the inter-series dependencies to enhance MTS forecasting. In this paper, we propose a Cluster-aware Attentive Convolutional Recurrent Network (CACRN) to capture both inter-series and intra-series dependencies in MTS data. Specifically, CACRN first introduces a cluster-aware variable representation module that separates irrelevant variables and captures the interaction between relevant variables to learn cluster-aware variable representations. Then, CACRN feeds these representations into parallel convolutional recurrent neural networks (CRNNs) to capture the short- and long-term temporal dependencies in a cluster-wise manner. Next, a cluster-aware attention mechanism is introduced to attend to temporal information in each cluster and co-attend all cluster information jointly to capture intra-cluster and inter-cluster dependencies for the downstream forecasting task. Our extensive experiments on six real-world datasets demonstrate that CACRN is effective and outperforms representative and state-of-the-art baselines. Our proposed method is suitable for a wide range of real-world data collections, especially those with clear dependencies of variables.

Rapid Chemical Screening of Microplastics and Nanoplastics by Thermal Desorption and Pyrolysis Mass Spectrometry with Unsupervised Fuzzy Clustering.

The transport and chemical identification of microplastics and nanoplastics (MNPs) are critical to the concerns over plastic accumulation in the environment. Chemically and physically transient MNP species present unique challenges for isolation and analysis due to many factors such as their size, color, surface properties, morphology, and potential for chemical change. These factors contribute to the eventual environmental and toxicological impact of MNPs. As analytical methods and instrumentation continue to be developed for this application, analytical test materials will play an important role. Here, a direct mass spectrometry screening method was developed to rapidly characterize manufactured and weathered MNPs, complementing lengthy pyrolysis-gas chromatography-mass spectrometry analysis. The chromatography-free measurements took advantage of Kendrick mass defect analysis, in-source collision-induced dissociation, and advancements in machine learning approaches for the data analysis of complex mass spectra. In this study, we applied Gaussian mixture models and fuzzy c-means clustering for the unsupervised analysis of MNP sample spectra, incorporating clustering stability and information criterion measurements to determine latent dimensionality. These models provided insight into the composition of mixed and weathered MNP samples. The multiparametric data acquisition and machine learning approach presented improved confidence in polymer identification and differentiation.

GOLFS: feature selection via combining both global and local information for high dimensional clustering

GOLFS: feature selection via combining both global and local information for high dimensional clustering

Quantifying Roman WFI Dark Images with the Wavelet Scattering Transform

The Nancy Grace Roman Space Telescope will survey a large area of the sky at near-infrared wavelengths with its Wide Field Instrument (WFI). The performance of the 18 WFI H4RG-10 detectors will need to be well-characterized and regularly monitored in order for Roman to meet its science objectives. Weak lensing science goals are particularly sensitive to instrumental distortions and patterns that might masquerade as astronomical signals. We apply the wavelet scattering transform in order to analyze localized signals in Roman WFI images that have been taken as part of a dark image test suite. The scattering transform quantifies shapes and clustering information by reducing images into nonlinear combinations of wavelet modes on multiple size scales. We show that these interpretable scattering statistics can separate rare correlated patterns from typical noise signals, and we discuss the results in context of power spectrum analyses and other computer vision methods.

A Riemannian Framework for Structurally Curated Functional Clustering of Brain White Matter Fibers.

White matter (WM) consists of fibers that transmit information from one brain region to another, and functional fiber clustering that combines diffusion and functional MRI provides a novel perspective for exploring the functional architecture of axonal fibers. However, existing methods only concern functional signals in gray matter (GM), whereas the connecting fibers may not transmit relevant functional signals. There has been growing evidence that neural activity is encoded in WM BOLD signals as well, which provides rich multimodal information for fiber clustering. In this paper, we develop a comprehensive Riemannian framework for functional fiber clustering using WM BOLD signals along fibers. Specifically, we derive a novel metric that is highly discriminative of different functional classes while reducing the variability within classes and, in the meantime, enables low-dimensional coding of high-dimensional data. Our in vivo experiments show that the proposed framework is able to achieve clustering results with inter-subject consistency and functional homogeneity. In addition, we develop an atlas of WM functional architecture for standardizable yet flexible use and exemplify a machine-learning-based application for the classification of autism spectrum disorders, which further demonstrates the great potential of our approach in practical applications.

The respective contribution of cognitive control and working memory to semantic and subjective organization in aging.

Organizing information is beneficial to episodic memory performance. Among several possible organizational strategies, two consist of organizing the information in semantic clusters (semantic organization) or self-organizing the information based on new associations that do not exist in semantic memory (subjective organization). Here, we investigated in a single study how these two organizational behaviors were underlined by different controlled processes and whether these relations were subjected to age-related differences. We tested 123 younger adults (n = 63) and older adults (n = 60) on two episodic memory tasks, one where the words were organizable and another where the words were not organizable, allowing for semantic and subjective organization, respectively. Additionally, participants were tested on three cognitive control tasks (Wisconsin Card Sorting Test, Stroop Test and Trail Making Test) and three working memory tasks (Backward Digit Span, Alpha Span and N-back test). Results revealed well-established age-related differences in terms of recall performance and organizational strategy implementation. More importantly, we found evidence that the different cognitive tests statistically yielded two different latent factors, a cognitive control factor and a working memory factor. Based on this dissociation, we found that only cognitive control contributed to semantic organization in all age groups whereas only working memory contributed to subjective organization, also in all age groups. These results shed new lights on our understanding of how controlled processes differently contribute to organizational behaviors in episodic memory. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Deep robust multi-channel learning subspace clustering networks

Subspace clustering methods are now widely used for unsupervised high-dimensional data processing in computer vision and other domains. Deep subspace clustering methods based on auto-encoder networks have made a significant improvement in nonlinear subspace clustering problems in comparison to previous works. However, these methods ignore the valid information lost during feature extraction, resulting in incomplete information and imprecise feature representations for subspace clustering. In addition, the clustering performance of the existing clustering methods is excessively dependent on hyper-parameters, making training difficult and unstable. In this paper, we propose Deep Robust Multi-Channel Learning Subspace Clustering Networks (DRMCLSC), a novel deep subspace clustering network for learning more comprehensive feature representations with good robustness for subspace clustering. The multi-channel learning strategy allows the model to extract, retain and fuse features simultaneously, enabling all valid information from the sample data to be obtained. Moreover, the multi-channel learning structure of the proposed method produces a more stable integration network that is less dependent on hyper-parameters and more resistant to training errors than previous works. Extensive experimental results on four benchmark datasets demonstrate the proposed method is superior and more effective than the state-of-the-art subspace clustering methods.

Low-rank discrete multi-view spectral clustering

Spectral clustering has attracted intensive attention in multimedia applications due to its good performance on arbitrary shaped clusters and well-defined mathematical framework. However, most existing multi-view spectral clustering methods still have the following demerits: (1) They ignore useful complementary information embedded in indicator matrices of different views. (2) The conventional post-processing methods based on the relax and discrete strategy inevitably result in the sub-optimal discrete solution. To tackle the aforementioned drawbacks, we propose a low-rank discrete multi-view spectral clustering model. Drawing inspiration from the fact that the difference between indicator matrices of different views provides useful complementary information for clustering, our model exploits the complementary information embedded in indicator matrices with tensor Schatten p-norm constraint. Further, we integrate low-rank tensor learning and discrete label recovering into a uniform framework, which avoids the uncertainty of the relaxed and discrete strategy. Extensive experiments on benchmark datasets have demonstrated the effectiveness and superiority of the proposed method.

Heterogeneous Graph Neural Network for identifying hadronically decayed tau leptons at the High Luminosity LHC

We present a new algorithm that identifies reconstructed jets originating from hadronic decays of tau leptons against those from quarks or gluons. No tau lepton reconstruction algorithm is used. Instead, the algorithm represents jets as heterogeneous graphs with tracks and energy clusters as nodes and trains a Graph Neural Network to identify tau jets from other jets. Different attributed graph representations and different GNN architectures are explored. We propose to use differential track and energy cluster information as node features and a heterogeneous sequentially-biased encoding for the inputs to final graph-level classification.

First Report of Mulberry Rust Caused by Cerotelium fici on Morus nigra in Brazil.

Leaf rust caused by Cerotelium fici (Cast.) Arth. is the main disease affecting Moraceae family plants, such as Ficus and Morus species (Galleti and Rezende 2016; Srikantaswamy et al. 2006). In August 2020, rust symptoms were observed in 100% of mulberry (Morus nigra L.) trees in an experimental orchard (Piracicaba, SP, Brazil; 22°42'28"S, 47°37'42"W). Mulberry leaves with high rust severity became yellowish and fell-off prematurely. Pustules were light brown with yellowish halo and presented mean size of 0.9 mm2. Uredinial paraphyses (n = 50) measured 42.2 ± 0.67 µm long with wall uniformly ca 0.6-1.1 μm thick. Urediniospores were brownish, echinulate, globoid to broadly ellipsoid, and measured 27.1 ± 0.29 × 21.0 ± 0.27 µm with a wall thickness of 0.6 ± 0.01 µm (n = 100). The morphology of the urediniospores observed in this study was similar to that reported in the literature for C. fici on Morus alba and Ficus spp. (Gupta et al. 1994; McKenzie 1986; Hennen et al. 2005). We used a low-coverage genome-skimming approach to retrieve genetic information of the rRNA cluster and the mtDNA. Genomic DNA was extracted from 3-4 mg of stored urediniospores at -80 °C, macerated in liquid nitrogen, using a modified cetyl trimethylammonium bromide extraction procedure (Lo Piccolo et al. 2012), and sequenced with 150-bp paired-end reads on Illumina NovaSeq 6000 System. Raw data, (45,761,957 X 2 reads) were assembled with SPAdes v3.15.1 (Bankevich et al., 2012) and the output used to create a custom BLAST database. Loci used for the phylogenetic analyses were identified by BLASTn using, as a query, sequences of C. fici from Ficus sp. from Australia publicly available: Accession No. MH047210.1 for the rRNA and MW036502.1 for COX3. The retrieved sequences were deposited in GenBank under accession numbers OM296992 and OP797407 for the partial rRNA cluster and COX3, respectively. The Bayesian inference phylogenetic analysis of the three concatenate loci (18S, 28S, and COX3) revealed that the isolate obtained in this study (MN1) was clustered in a well-supported clade with C. fici type species. Pathogenicity tests were conducted using mulberry potted plants under greenhouse conditions (25 ± 5 °C). The urediniospores suspension (5 × 104 urediniospores ml-1) with 0.05% Tween 20 was sprayed with an airbrush on fully expanded leaves until run-off. As a control, mulberry plants were sprayed with distilled water and kept under the same conditions. Inoculated and mock-inoculated plants were kept in a dark moist chamber at 23 °C (± 2 °C) for 24 h. After this period, plants were moved to the greenhouse. The experimental design was completely randomized with five replicates, each replicate consisted of one potted plant and the experiment was performed twice. At 12 days post-inoculation, all inoculated plants showed rust symptoms identical to those observed in the field, whereas control plants had no symptoms. The first symptoms were small pustules on the abaxial surface of fully expanded leaves. Small chlorotic lesions were observed on the adaxial leaf surface, which evolved into necrotic lesions. The pathogen was re-inoculated into potted plants, where it was maintained through monthly inoculations. To our knowledge, this is the first report of mulberry rust on M. nigra in Brazil. As mulberry leaves are the only natural food for silkworm (Bombyx mori L.), rust poses a significant threat to the sericulture industry because the disease can decrease production and quality of mulberry foliage.

Cluster-Guided Contrastive Graph Clustering Network

Benefiting from the intrinsic supervision information exploitation capability, contrastive learning has achieved promising performance in the field of deep graph clustering recently. However, we observe that two drawbacks of the positive and negative sample construction mechanisms limit the performance of existing algorithms from further improvement. 1) The quality of positive samples heavily depends on the carefully designed data augmentations, while inappropriate data augmentations would easily lead to the semantic drift and indiscriminative positive samples. 2) The constructed negative samples are not reliable for ignoring important clustering information. To solve these problems, we propose a Cluster-guided Contrastive deep Graph Clustering network (CCGC) by mining the intrinsic supervision information in the high-confidence clustering results. Specifically, instead of conducting complex node or edge perturbation, we construct two views of the graph by designing special Siamese encoders whose weights are not shared between the sibling sub-networks. Then, guided by the high-confidence clustering information, we carefully select and construct the positive samples from the same high-confidence cluster in two views. Moreover, to construct semantic meaningful negative sample pairs, we regard the centers of different high-confidence clusters as negative samples, thus improving the discriminative capability and reliability of the constructed sample pairs. Lastly, we design an objective function to pull close the samples from the same cluster while pushing away those from other clusters by maximizing and minimizing the cross-view cosine similarity between positive and negative samples. Extensive experimental results on six datasets demonstrate the effectiveness of CCGC compared with the existing state-of-the-art algorithms. The code of CCGC is available at https://github.com/xihongyang1999/CCGC on Github.

Hard Sample Aware Network for Contrastive Deep Graph Clustering

Contrastive deep graph clustering, which aims to divide nodes into disjoint groups via contrastive mechanisms, is a challenging research spot. Among the recent works, hard sample mining-based algorithms have achieved great attention for their promising performance. However, we find that the existing hard sample mining methods have two problems as follows. 1) In the hardness measurement, the important structural information is overlooked for similarity calculation, degrading the representativeness of the selected hard negative samples. 2) Previous works merely focus on the hard negative sample pairs while neglecting the hard positive sample pairs. Nevertheless, samples within the same cluster but with low similarity should also be carefully learned. To solve the problems, we propose a novel contrastive deep graph clustering method dubbed Hard Sample Aware Network (HSAN) by introducing a comprehensive similarity measure criterion and a general dynamic sample weighing strategy. Concretely, in our algorithm, the similarities between samples are calculated by considering both the attribute embeddings and the structure embeddings, better revealing sample relationships and assisting hardness measurement. Moreover, under the guidance of the carefully collected high-confidence clustering information, our proposed weight modulating function will first recognize the positive and negative samples and then dynamically up-weight the hard sample pairs while down-weighting the easy ones. In this way, our method can mine not only the hard negative samples but also the hard positive sample, thus improving the discriminative capability of the samples further. Extensive experiments and analyses demonstrate the superiority and effectiveness of our proposed method. The source code of HSAN is shared at https://github.com/yueliu1999/HSAN and a collection (papers, codes and, datasets) of deep graph clustering is shared at https://github.com/yueliu1999/Awesome-Deep-Graph-Clustering on Github.

ReporTree: a surveillance-oriented tool to strengthen the linkage between pathogen genetic clusters and epidemiological data

BackgroundGenomics-informed pathogen surveillance strengthens public health decision-making, playing an important role in infectious diseases’ prevention and control. A pivotal outcome of genomics surveillance is the identification of pathogen genetic clusters and their characterization in terms of geotemporal spread or linkage to clinical and demographic data. This task often consists of the visual exploration of (large) phylogenetic trees and associated metadata, being time-consuming and difficult to reproduce.ResultsWe developed ReporTree, a flexible bioinformatics pipeline that allows diving into the complexity of pathogen diversity to rapidly identify genetic clusters at any (or all) distance threshold(s) or cluster stability regions and to generate surveillance-oriented reports based on the available metadata, such as timespan, geography, or vaccination/clinical status. ReporTree is able to maintain cluster nomenclature in subsequent analyses and to generate a nomenclature code combining cluster information at different hierarchical levels, thus facilitating the active surveillance of clusters of interest. By handling several input formats and clustering methods, ReporTree is applicable to multiple pathogens, constituting a flexible resource that can be smoothly deployed in routine surveillance bioinformatics workflows with negligible computational and time costs. This is demonstrated through a comprehensive benchmarking of (i) the cg/wgMLST workflow with large datasets of four foodborne bacterial pathogens and (ii) the alignment-based SNP workflow with a large dataset of Mycobacterium tuberculosis. To further validate this tool, we reproduced a previous large-scale study on Neisseria gonorrhoeae, demonstrating how ReporTree is able to rapidly identify the main species genogroups and characterize them with key surveillance metadata, such as antibiotic resistance data. By providing examples for SARS-CoV-2 and the foodborne bacterial pathogen Listeria monocytogenes, we show how this tool is currently a useful asset in genomics-informed routine surveillance and outbreak detection of a wide variety of species.ConclusionsIn summary, ReporTree is a pan-pathogen tool for automated and reproducible identification and characterization of genetic clusters that contributes to a sustainable and efficient public health genomics-informed pathogen surveillance. ReporTree is implemented in python 3.8 and is freely available at https://github.com/insapathogenomics/ReporTree.

Signatures of Neuropsychological Test Results in the Long Life Family Study: A Cluster Analysis.

Discovering patterns of cognitive domains and characterizing how these patterns associate with other risk factors and biomarkers can improve our understanding of the determinants of cognitive aging. To discover patterns of cognitive domains using neuropsychological test results in Long Life Family Study (LLFS) and characterize how these patterns associate with aging markers. 5,086 LLFS participants were administered neuropsychological tests at enrollment. We performed a cluster analysis of six baseline neuropsychological test scores and tested the association between the identified clusters and various clinical variables, biomarkers, and polygenic risk scores using generalized estimating equations and the Chi-square test. We used Cox regression to correlate the clusters with the hazard of various medical events. We investigated whether the cluster information could enhance the prediction of cognitive decline using Bayesian beta regression. We identified 12 clusters with different cognitive signatures that represent profiles of performance across multiple neuropsychological tests. These signatures significantly correlated with 26 variables including polygenic risk scores, physical and pulmonary functions, and blood biomarkers and were associated with the hazard of mortality (p < 0.01), cardiovascular disease (p = 0.03), dementia (p = 0.01), and skin cancer (p = 0.03). The identified cognitive signatures capture multiple domains simultaneously and provide a holistic vision of cognitive function, showing that different patterns of cognitive function can coexist in aging individuals. Such patterns can be used for clinical intervention and primary care.

Joint Slot Scheduling and Power Allocation in Clustered Underwater Acoustic Sensor Networks

Dear Editor This letter deals with the joint slot scheduling and power allocation in clustered underwater acoustic sensor networks (UASNs), based on the known clustering and routing information, to maximize the network's energy efficiency (EE). Based on the block coordinated decent (BCD) method, the formulated mixed-integer non-convex problem is alternatively optimized by leveraging the Kuhn-Munkres algorithm, the Dinkelbach's method and the successive convex approximation (SCA) technique. Numerical results show that the proposed scheme has a better performance in maximizing EE compared to the separate optimization methods.