Clustering Approach Research Articles

First Ab Initio Line Lists for Triatomic Sulfur-Containing Molecules: S2O and S3.

We present in this paper the first room temperature rovibrational line lists for the main isotopologues of disulfur monoxide (S2O) and thiozone (S3) variationally computed from newly developed ab initio potential energy and dipole moment surfaces (PESs and DMSs). S2O and S3 are supposed to be potential candidates for astronomical detection, especially in the Venusian atmosphere where sulfur chemistry plays a major role. Contrary to other stable sulfur-containing species like SO2 or H2S, there is much less experimental data for the short-lived reactive S2O and S3 molecules. Indeed, the infrared spectra of S2O and S3 are quite complicated to analyze without consistent theoretical predictions because they contain both high J (∼100) and a lot of hot band transitions, even at T = 296 K. In this work, we have constructed PESs based on the single-reference coupled cluster approach [CCSD(T)] and including corrections due to the scalar relativistic effects, DBOC, and highly excited Slater determinants. The structure of the excited electronic states of S3 was also discussed. The nuclear-motion Eckart-Watson Hamiltonian expressed in terms of normal-mode irreducible tensor operators was used to compute the energy levels. For line intensity calculation, ab initio DMSs were computed at the CCSD(T)/aug-cc-pV5Z level of the theory. Rotation-vibration patterns in the fundamental bands of S2O and S3 have been simulated taking into account the recent Fourier-transform spectra of S2O recorded at the SOLEIL synchrotron. The present work aims at providing line intensities of S2O and S3 that are missing in the literature as well as new spectroscopic support for atmospheric applications.

An efficient network clustering approach using graph-boosting and nonnegative matrix factorization

Network clustering is a critical task in data analysis, aimed at uncovering the underlying structure and patterns within complex networks. Traditional clustering methods often struggle with large-scale and noisy data, leading to suboptimal results. Also, the efficiency of positive samples in network clustering depends on the carefully constructed data augmentation, and the pre-training process of the model deals with large-scale data. To address these issues, in this paper, we introduce an efficient network clustering approach that leverages Graph-Boosting and Nonnegative Matrix Factorization to enhance clustering performance (GBNMF). Our algorithm addresses the limitations of traditional clustering techniques by incorporating the strengths of graph-boosting, which iteratively improves the quality of clusters, and Nonnegative Matrix Factorization (NMF), which effectively captures latent structures within the data. We validate our algorithm through extensive experiments on various benchmark network datasets, demonstrating significant improvements in clustering accuracy and robustness. The proposed algorithm not only achieves superior clustering results but also exhibits remarkable computational efficiency, making it a valuable tool for large-scale network analysis applications.

Identifying heterogeneous subgroups of systemic autoimmune diseases by applying a joint dimension reduction and clustering approach to immunomarkers

BackgroundThe high complexity of systemic autoimmune diseases (SADs) has hindered precise management. This study aims to investigate heterogeneity in SADs.MethodsWe applied a joint cluster analysis, which jointed multiple correspondence analysis and k-means, to immunomarkers and measured the heterogeneity of clusters by examining differences in immunomarkers and clinical manifestations. The electronic health records of patients who received an antinuclear antibody test and were diagnosed with SADs, namely systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), and Sjögren’s syndrome (SS), were retrieved between 2001 and 2016 from hospitals in Taiwan.ResultsWith distinctive patterns of immunomarkers, a total of 11,923 patients with the three SADs were grouped into six clusters. None of the clusters was composed only of a single SAD, and these clusters demonstrated considerable differences in clinical manifestation. Both patients with SLE and SS had a more dispersed distribution in the six clusters. Among patients with SLE, the occurrence of renal compromise was higher in Clusters 3 and 6 (52% and 51%) than in the other clusters (p < 0.001). Cluster 3 also had a high proportion of patients with discoid lupus (60%) than did Cluster 6 (39%; p < 0.001). Patients with SS in Cluster 3 were the most distinctive because of the high occurrence of immunity disorders (63%) and other and unspecified benign neoplasm (58%) with statistical significance compared with the other clusters (all p < 0.05).ConclusionsThe immunomarker-driven clustering method could recognise more clinically relevant subgroups of the SADs and would provide a more precise diagnosis basis.

Quantifying Treatments as Usual and with Technologies in Neurorehabilitation of Individuals with Spinal Cord Injury.

Several technologies have been introduced into neurorehabilitation programs to enhance traditional treatment of individuals with Spinal Cord Injury (SCI). Their effectiveness has been widely investigated, but their adoption has not been properly quantified. The aim of this study is to assess the distribution of conventional (Treatment As Usual-TAU) and technology-aided (Treatment With Technologies-TWT) treatments conveniently grouped based on different therapeutic goals in a selected SCI unit. Data from 104 individuals collected in 29 months were collected in a custom database and categorized according to both the conventional American Impairment Scale classification and a newly developed Multifactor (MF) clustering approach that considers additional sources of information (the lesion level, the level of independence in the activities of daily living, and the hospitalization duration). Results indicated an average technology adoption of about 30%. Moreover, the MF clusters were less overlapped, and the differences in TWT adoption were more pronounced than in AIS-based clustering. MF clustering was capable of grouping individuals based both on neurological features and functional abilities. In particular, individuals with motor complete injuries were grouped together, whereas individuals with sensorimotor incomplete SCI were collected separately based on the lesion level. As regards TWT adoption, we found that in the case of motor complete SCI, TWT for muscle tone control and modulation was mainly selected (about 90% of TWT), while the other types of TWT were seldom adopted. Even for individuals with incomplete SCI, the most frequent rehabilitation goal was muscle tone modulation (about 75% of TWT), regardless of the AIS level, and technologies to improve walking ability (about 12% of TWT) and balance control (about 10% of TWT) were mainly used for individuals with thoracic or lumbar lesions. Analyzing TAU distribution, we found that the highest adoption of muscle tone modulation strategies was reported in the case of individuals with motor complete SCI (about 42% of TAU), that is, in cases when almost no gait training was pursued (about 1% of TAU). In the case of cervical motor incomplete SCI, compared to thoracic and lumbar incomplete SCI, there was a greater focus on muscle tone control and force recruitment in addition to walking training (38% and 14% of TAU, respectively) than on balance training. Overall, the MF clustering provided more insights than the traditional AIS-based classification, highlighting differences in TWT adoption. These findings suggest that a wider overview that considers both neurological and functional characteristics of individuals after SCI based on a multifactor analysis could enhance the personalization of neurorehabilitation strategies.

Enhancing VANET communication using squid game optimization based energy aware clustering approach

Enhancing VANET communication using squid game optimization based energy aware clustering approach

Integrating UAV imagery and machine learning via Geographic Object Based Image Analysis (GEOBIA) for enhanced monitoring of Yucca gloriosa in Mediterranean coastal dunes

Effective monitoring and early detection of invasive alien plant species (IAPs) are crucial for mitigating their spread and safeguarding native habitats. Unmanned Aerial Vehicles (UAVs) offer a cost-efficient solution, providing high resolution images. In this study, we aimed to develop a semi-automated methodology using a machine learning algorithm, spatial metrics, and clustering techniques on UAV images to monitor, map, and suggest management measures to counteract Yucca gloriosa, an invasive plant colonizing coastal fixed dunes in central Italy.UAV flights were conducted using two drones: one for the visible spectrum and the other for multispectral bands (Blue, Green, Red, Red Edge, and Near Infrared) along with a Digital Surface Model (DSM). Derived vegetation indices were also utilized. For mapping Y. gloriosa distribution, a Geographic Object Based Image Analysis (GEOBIA) approach was applied to the orthophoto segmentation, followed by a Random Forest algorithm in a training phase, considering three variable combinations (DSM + vegetation indices, DSM + spectral bands, DSM + mixed variables). The most accurate Y. gloriosa map was used to suggest management measures combining the spatial pattern of invaded patches (size, height, isolation level, and aggregation degree) and a mixed clustering approach (hierarchical and partitioning).The results highlighted that the most accurate prediction map was based on the DSM + mixed variables dataset, showing the important role of using a combination of spectral bands and vegetation indices. In all three cases, the DSM emerged as the pivotal variable for discriminating Y. gloriosa from the surrounding environment. Additionally, our results demonstrate the advantages of incorporating vegetation indices in discerning the target invasive alien plant (IAP) from the broader environment, particularly considering its distinctive photosynthesis process and biomass production. From a managerial standpoint, our pilot study indicates that the UAV-based mapping methodology represents an optimal balance between field efforts and costs. This approach allows for the precise identification of containment and removal areas of Y. gloriosa, without compromising the accuracy of the method. The generated prediction maps also hold potential significance for the conservation of coastal dune ecosystems, providing a promising tool for the effective management of invasive species and biodiversity conservation by suggesting management measures for Y. gloriosa.

Exploring meta-heuristics for partitional clustering: methods, metrics, datasets, and challenges

Partitional clustering is a type of clustering that can organize the data into non-overlapping groups or clusters. This technique has diverse applications across the different various domains like image processing, pattern recognition, data mining, rule-based systems, customer segmentation, image segmentation, and anomaly detection, etc. Hence, this survey aims to identify the key concepts and approaches in partitional clustering. Further, it also highlights its widespread applicability including major advantages and challenges. Partitional clustering faces challenges like selecting the optimal number of clusters, local optima, sensitivity to initial centroids, etc. Therefore, this survey describes the clustering problems as partitional clustering, dynamic clustering, automatic clustering, and fuzzy clustering. The objective of this survey is to identify the meta-heuristic algorithms for the aforementioned clustering. Further, the meta-heuristic algorithms are also categorised into simple meta-heuristic algorithms, improved meta-heuristic algorithms, and hybrid meta-heuristic algorithms. Hence, this work also focuses on the adoption of new meta-heuristic algorithms, improving existing methods and novel techniques that enhance clustering performance and robustness, making partitional clustering a critical tool for data analysis and machine learning. This survey also highlights the different objective functions and benchmark datasets adopted for measuring the effectiveness of clustering algorithms. Before the literature survey, several research questions are formulated to ensure the effectiveness and efficiency of the survey such as what are the various meta-heuristic techniques available for clustering problems? How to handle automatic data clustering? What are the main reasons for hybridizing clustering algorithms? The survey identifies shortcomings associated with existing algorithms and clustering problems and highlights the active area of research in the clustering field to overcome these limitations and improve performance.

Harnessing machine learning algorithms to unveil energy efficiency investment archetypes

Increasing transparency about the performance of different projects is crucial to reducing the heterogeneity in the energy efficiency services market, thereby upscaling investments. In this context, machine learning algorithms could assist in identifying and analyzing energy efficiency project archetypes, although this field has so far been explored with a limited view in the literature. This paper aims to address this gap by identifying energy efficiency investment families and the determinant factors of the classification scheme, using machine learning. In this effort, it hinges on a wide range of indicators from implemented projects around Europe and the USA, including investment profitability, initial investment, risk of failure, intervention type, life measure, region of implementation and building type. The analysis employs two clustering approaches, namely Partitioning Around Medoids (PAM) and K-means, determining the number of clusters based on the Silhouette index and total within–cluster sum of squares. The results indicate that energy efficiency investments can be classified into three categories: (i) “junk investments”, characterized by low–profitability (IRR∼10%), moderate risk, and extended horizons; (ii) “safe profitability”, distinguished by high profitability (IRR∼30%) and minimal risk; and (iii) “high stakes”, described by exceptionally high profitability (IRR∼40%), coupled with a substantial risk. Next to profitability and risk of failure, also energy efficiency intervention and building type (sector) emerge among the most influential factors in the classification scheme. Feature importance shows a significant sensitivity to the chosen classification model.

RETRACTED ARTICLE: A clustering approach for attack detection and data transmission in vehicular ad-hoc networks

ABSTRACT We, the Editors and Publisher of Network: Computation in Neural Systems, have retracted the following article: Barve, A., & Patheja, P. S. (2023). A clustering approach for attack detection and data transmission in vehicular ad-hoc networks. Network: Computation in Neural Systems, 1–26. https://doi.org/10.1080/0954898X.2023.2279973 Since publication, significant concerns have been raised about the fact that this article has substantial overlaps with the following article: Barve, A. & Patheja, P. S. (2023). A clustering approach for attack detection and data transmission in vehicular ad-hoc networks. Ad Hoc & Sensor Wireless Networks, 58. 1-2, p. 127-149. DOI: 10.32908/ahswn.v58.10375 Further investigations by the Publisher revealed that these overlaps are present in all sections of the article, including the figures and tables without appropriate acknowledgement. Upon query, the authors agree that the article is a duplicate submission. As this is a serious breach of our Editorial Policies, we are retracting the article from the journal. The corresponding author listed in this publication has been informed. We have been informed in our decision-making by our editorial policies and the COPE guidelines. The retracted article will remain online to maintain the scholarly record, but it will be digitally watermarked on each page as ‘Retracted’.

Optimizing Local Climate Zones through Clustering for Surface Urban Heat Island Analysis in Building Height-Scarce Cities: A Cape Town Case Study

Studying air Urban Heat Islands (AUHI) in African cities is limited by building height data scarcity and sparse air temperature (Tair) networks, leading to classification confusion and gaps in Tair data. Satellite imagery used in surface UHI (SUHI) applications overcomes the gaps which befall AUHI, thus making it the primary focus of UHI studies in areas with limited Tair stations. Consequently, we used Landsat 30 m imagery to analyse SUHI patterns using Land Surface Temperature (LST) data. Local climate zones (LCZ) as a UHI study tool have been documented to not result in distinct thermal environments at the surface level per LCZ class. The goal in this study was thus to explore relationships between LCZs and LST patterns, aiming to create a building height (BH)-independent LCZ framework capable of creating distinct thermal environments to study SUHI in African cities where LiDAR data are scarce. Random forests (RF) classified LCZ in R, and the Single Channel Algorithm (SCA) extracted LST via the Google Earth Engine. Statistical analyses, including ANOVA and Tukey’s HSD, assessed thermal distinctiveness, using a 95% confidence interval and 1 °C threshold for practical significance. Semi-Automated Agglomerative Clustering (SAAC) and Automated Divisive Clustering (ADC) grouped LCZs into thermally distinct clusters based on physical characteristics and LST data internal patterns. Built LCZs (1–9) had higher mean LSTs; LCZ 8 reached 37.6 °C in Spring, with a smaller interquartile range (IQR) (34–36 °C) and standard deviation (SD) (1.85 °C), compared to natural classes (A–G) with LCZ 11 (A–B) at 14.9 °C/LST, 17–25 °C/IQR, and 4.2 °C SD. Compact LCZs (2, 3) and open LCZs (5, 6), as well as similar LCZs in composition and density, did not show distinct thermal environments even with building height included. The SAAC and ADC clustered the 14 LCZs into six thermally distinct clusters, with the smallest LST difference being 1.19 °C, above the 1 °C threshold. This clustering approach provides an optimal LCZ framework for SUHI studies, transferable to different urban areas without relying on BH, making it more suitable than the full LCZ typology, particularly for the African context. This clustered framework ensures a thermal distinction between clusters large enough to have practical significance, which is more useful in urban planning than statistical significance.

Automatic deep sparse clustering with a dynamic population-based evolutionary algorithm using reinforcement learning and transfer learning

Clustering data effectively remains a significant challenge in machine learning, particularly when the optimal number of clusters is unknown. Traditional deep clustering methods often struggle with balancing local and global search, leading to premature convergence and inefficiency. To address these issues, we introduce ADSC-DPE-RT (Automatic Deep Sparse Clustering with a Dynamic Population-based Evolutionary Algorithm using Reinforcement Learning and Transfer Learning), a novel deep clustering approach. ADSC-DPE-RT builds on Multi-Trial Vector-based Differential Evolution (MTDE), an algorithm that integrates sparse auto-encoding and manifold learning to enable automatic clustering without prior knowledge of cluster count. However, MTDE's fixed population size can lead to either prolonged computation or premature convergence. Our approach introduces a dynamic population generation technique guided by Reinforcement Learning (RL) and Markov Decision Process (MDP) principles. This allows for flexible adjustment of population size, preventing premature convergence and reducing computation time. Additionally, we incorporate Generative Adversarial Networks (GANs) to facilitate dynamic knowledge transfer between MTDE strategies, enhancing diversity and accelerating convergence towards the global optimum. This is the first work to address the dynamic population issue in deep clustering through RL, combined with Transfer Learning to optimize evolutionary algorithms. Our results demonstrate significant improvements in clustering performance, positioning ADSC-DPE-RT as a competitive alternative to state-of-the-art deep clustering methods.

Integration of Hydrogen-Deuterium Exchange Mass Spectrometry with Molecular Dynamics Simulations and Ensemble Reweighting Enables High Resolution Protein-Ligand Modeling.

Hydrogen-Deuterium exchange mass spectrometry's (HDX-MS) utility in identifying and characterizing protein-small molecule interaction sites has been established. The regions that are seen to be protected from exchange upon ligand binding indicate regions that may be interacting with the ligand, giving a qualitative understanding of the ligand binding pocket. However, quantitatively deriving an accurate high-resolution structure of the protein-ligand complex from the HDX-MS data remains a challenge, often limiting its use in applications such as small molecule drug design. Recent efforts have focused on the development of methods to quantitatively model Hydrogen-Deuterium exchange (HDX) data from computationally modeled structures to garner atomic level insights from peptide-level resolution HDX-MS. One such method, HDX ensemble reweighting (HDXer), employs maximum entropy reweighting of simulated HDX data to experimental HDX-MS to model structural ensembles. In this study, we implement and validate a workflow which quantitatively leverages HDX-MS data to accurately model protein-small molecule ligand interactions. To that end, we employ a strategy combining computational protein-ligand docking, molecular dynamics simulations, HDXer, and dimensional reduction and clustering approaches to extract high-resolution drug binding poses that most accurately conform with HDX-MS data. We apply this workflow to model the interaction of ERK2 and FosA with small molecule compounds and inhibitors they are known to bind. In five out of six of the protein-ligand pairs tested, the HDX derived protein-ligand complexes result in a ligand root-mean-square deviation (RMSD) within 2.5 Å of the known crystal structure ligand.

Dense SNP-based analyses complement forensic anthropology biogeographical ancestry assessments

Identification of unidentified human remains (UHRs) is crucial yet challenging, especially with traditional forensic techniques. Forensic anthropological examinations can yield ancestry estimations; however, the utility of these estimates is limited by the data points that can be collected from partial remains, complexities of admixture, and variation of phenotypic expression due to environmental effects. While it is generally known that anthropological estimates can be imprecise, the performance of these methods has not been studied at scale. Genome-wide SNP testing is an orthogonal approach for estimating ancestry and offers a unique opportunity to measure the magnitude of anthropological ancestry misattribution. Genomic ancestry inference leverages principal component analysis (PCA) and model-based clustering approaches. This study compares anthropologically determined ancestry with those estimated using genome-wide SNP markers. A dataset of 611 UHR samples with publicly available ancestry assessments from National Missing and Unidentified Persons System (NamUs) was analyzed. The genetic ancestry approach, validated against reference population samples, offers robust ancestry calculations for major population groups. Inconsistency between anthropological and genomic ancestry assignments were observed, particularly for admixed populations. Although forensic anthropological examinations remain valuable, their limitations emphasize the need for refinement and enhancement through the augmentation of SNP-based analyses. Further validation studies are crucial to define the uncertainty associated with both anthropological and genome-based ancestry estimates to resolve cases and aid law enforcement investigations. Additionally, current policy and practices for reporting ancestry for UHRs should be revisited to reduce potential misinformation.

Application of compound poisson model to estimate underreported risk of non-communicable diseases in underdeveloped areas

BackgroundHypertension and diabetes are major components of non-communicable diseases (NCDs), with a substantial number of patients residing in underdeveloped areas. Limited medical resources in these areas often results in underreporting of disease prevalence, masking the true extent of diseases. Taking the underdeveloped Liangshan Yi Autonomous Prefecture in China as an example, this study aimed to correct the underreported prevalence of hypertension and type 2 diabetes so as to provide inspiration for the allocation of medical resources in such areas. MethodsAssuming the true number of patients in each area follows a Poisson distribution, we applied a Compound Poisson Model based on Clustering of Data Quality (CPM-CDQ) to estimate the potential true prevalence of hypertension and diabetes, as well as the registration rate of existing patients. Specifically, a hierarchical clustering approach was utilized to group the counties based on the data quality, and then the registration rate of the cluster with the best data quality was used as a priori information for the model. The model parameters were estimated by the maximum likelihood method. Sensitivity analyses were performed to test the robustness of the model. ResultsThe estimated prevalence of hypertension in the entire Liangshan Prefecture from 2018 to 2020 ranged from 24.59 % to 25.28 %, and for diabetes, it ranged from 4.95 % to 8.42 %. The registration rates for hypertension and diabetes were 14.10 % to 24.59 % and 15.98 % to 29.12 %, respectively. Additionally, the accuracy of clustering the counties with the best data quality had a significant impact on the performance of the model. ConclusionLiangshan Prefecture is experiencing a significant high prevalence of hypertension and diabetes, accompanied by a concerningly low registration rate. The CPM-CDQ proved useful for assessing underreporting risks and facilitating targeted interventions for NCDs control and prevention, particularly in underdeveloped areas.

Retraction: A clustering approach for attack detection and data transmission in vehicular ad-hoc networks

Retraction: A clustering approach for attack detection and data transmission in vehicular ad-hoc networks

Scrutinizing Smartphone-Induced Stressors: A Clustering Approach

Scrutinizing Smartphone-Induced Stressors: A Clustering Approach

AutoFocus: a hierarchical framework to explore multi-omic disease associations spanning multiple scales of biomolecular interaction

Recent advances in high-throughput measurement technologies have enabled the analysis of molecular perturbations associated with disease phenotypes at the multi-omic level. Such perturbations can range in scale from fluctuations of individual molecules to entire biological pathways. Data-driven clustering algorithms have long been used to group interactions into interpretable functional modules; however, these modules are typically constrained to a fixed size or statistical cutoff. Furthermore, modules are often analyzed independently of their broader biological context. Consequently, such clustering approaches limit the ability to explore functional module associations with disease phenotypes across multiple scales. Here, we introduce AutoFocus, a data-driven method that hierarchically organizes biomolecules and tests for phenotype enrichment at every level within the hierarchy. As a result, the method allows disease-associated modules to emerge at any scale. We evaluated this approach using two datasets: First, we explored associations of biomolecules from the multi-omic QMDiab dataset (n = 388) with the well-characterized type 2 diabetes phenotype. Secondly, we utilized the ROS/MAP Alzheimer’s disease dataset (n = 500), consisting of high-throughput measurements of brain tissue to explore modules associated with multiple Alzheimer’s Disease-related phenotypes. Our method identifies modules that are multi-omic, span multiple pathways, and vary in size. We provide an interactive tool to explore this hierarchy at different levels and probe enriched modules, empowering users to examine the full hierarchy, delve into biomolecular drivers of disease phenotype within a module, and incorporate functional annotations.

A hybrid genetic-fuzzy ant colony optimization algorithm for automatic K-means clustering in urban global positioning system

This paper introduces an innovative automatic K-means clustering algorithm, namely HGA-FACO, which seamlessly integrates the noise algorithm, Genetic Algorithm (GA), Ant Colony Optimization (ACO), and Adaptive Fuzzy System (AFS). The rationale behind the HGA-FACO algorithm is to mitigate the shortcomings of traditional K-means, particularly the reliance on pre-determined cluster centers and the need for specifying the number of clusters in advance. By optimizing the search strategy, HGA-FACO efficiently circumvents local optima and effectively explores the global optimal solution, resulting in more accurate and stable clustering outcomes. To validate the superiority of the HGA-FACO over conventional K-Means Clustering (KMeans) and other intelligent clustering approaches such as ACO-KMeans, GA-KMeans (GAK), particle swarm optimization KMeans (PSOK), and ACO-GAK, we conducted comprehensive experiments on taxi Global Positioning System (GPS) datasets sourced from four distinct cities. Employing rigorous evaluation metrics including Silhouette Coefficient (SC), Partition Coefficient (PBM), Davies-Bouldin Index (DBI), and Sum of Squared Errors (SSE), the experimental results convincingly demonstrate that the HGA-FACO significantly outperforms its counterparts across all metrics, highlighting its exceptional performance in clustering effectiveness and compactness. While the HGA-FACO faces challenges related to computational complexity and the necessity for initial parameter tuning, its performance limitations on small-sized or unevenly distributed datasets are acknowledged. Nevertheless, the algorithm's advancements in the field of clustering algorithms are undeniable and hold immense potential for practical applications, notably in city hotspot identification.

Learning clustering-friendly representations via partial information discrimination and cross-level interaction

Despite significant advances in the deep clustering research, there remain three critical limitations to most of the existing approaches. First, they often derive the clustering result by associating some distribution-based loss to specific network layers, neglecting the potential benefits of leveraging the contrastive sample-wise relationships. Second, they frequently focus on representation learning at the full-image scale, overlooking the discriminative information latent in partial image regions. Third, although some prior studies perform the learning process at multiple levels, they mostly lack the ability to exploit the interaction between different learning levels. To overcome these limitations, this paper presents a novel deep image clustering approach via Partial Information discrimination and Cross-level Interaction (PICI). Specifically, we utilize a Transformer encoder as the backbone, coupled with two types of augmentations to formulate two parallel views. The augmented samples, integrated with masked patches, are processed through the Transformer encoder to produce the class tokens. Subsequently, three partial information learning modules are jointly enforced, namely, the partial information self-discrimination (PISD) module for masked image reconstruction, the partial information contrastive discrimination (PICD) module for the simultaneous instance- and cluster-level contrastive learning, and the cross-level interaction (CLI) module to ensure the consistency across different learning levels. Through this unified formulation, our PICI approach for the first time, to our knowledge, bridges the gap between the masked image modeling and the deep contrastive clustering, offering a novel pathway for enhanced representation learning and clustering. Experimental results across six image datasets demonstrate the superiority of our PICI approach over the state-of-the-art. In particular, our approach achieves an ACC of 0.772 (0.634) on the RSOD (UC-Merced) dataset, which shows an improvement of 29.7% (24.8%) over the best baseline. The source code is available at https://github.com/Regan-Zhang/PICI.

K-medoid clustering containerized allocation algorithm for cloud computing environment

Load balancing is critical for container-based cloud computing environments for several reasons. A lack of appropriate load balancing techniques could result in a decrease in performance and possible service interruptions as some nodes get overloaded, while others are left underutilized. Cloud service providers can reduce latency and boost system performance by strategically placing containers using clustering algorithms. These techniques aid in efficiently using resources and load balancing by clustering related containers together according to their shared attributes. Clustering strategies are effective in allocating and controlling resources to meet the demands of a changing workload. Algorithms for clustering combine related workloads or containers into clusters, improving performance isolation and maximizing resource usage. One popular methodology for data clustering is the K-Medoid Clustering Algorithm. It is especially helpful when working with categorical data or when the dataset contains outliers. K-medoids is an unsupervised clustering approach where the core of the cluster is made up of data points known as “medoids.” A medoid is a location in the cluster whose total distance to every object in the cluster—also known as its dissimilarity—is as small as possible. Any appropriate distance function may be used, such as the Manhattan distance, the Euclidean distance, or another one. Thus, by choosing K medoids from our data sample, the K-medoids method splits the data into K clusters. This work presents the K-Medoid clustering technique for containers, which can enhance load balancing, decrease resource execution times, and increase resource utilization rates all at the same time. The results of the experiment show that the proposed method outperforms MACO and FCFS in terms of throughput by about 70% when number of cloudlets increased. The relative improvement of execution time of the proposed K-medoid algorithm w.r.t FCFS is about 50%.