Computational Clustering Research Articles

Molecular structure of 3,4-dicyanofuroxan studied by gas electron diffraction. Application of DFT theory and coupled cluster computations

Equilibrium molecular structure of 3,4-dicyanofuroxan has been determined for the first time by means of gas electron diffraction (GED) and quantum chemical (QC) calculations. The GED values have been compared to those obtained from QC computations at different levels of theory. It has been discovered that the value of the O1−N2 bond is very sensitive to the QC method applied. The best fit to the experimental data has performed CCSD(T)/cc-pVTZ, whereas DFT methods applied in this work have been found to be less accurate. Peculiar features of 3,4-dicyanofuroxan molecular structure have been revealed by applying NBO, AIM and NCI analyses.

DQN-based intelligent controller for multiple edge domains

Advanced technologies like network function virtualization (NFV) and multi-access edge computing (MEC) have been used to build flexible, highly programmable, and autonomously manageable infrastructures close to the end-users, at the edge of the network. In this vein, the use of single-board computers (SBCs) in commodity clusters has gained attention to deploy virtual network functions (VNFs) due to their low cost, low energy consumption, and easy programmability. This paper deals with the problem of deploying VNFs in a multi-cluster system formed by this kind of node which is characterized by limited computational and battery capacities. Additionally, existing platforms to orchestrate and manage VNFs do not consider energy levels during their placement decisions, and therefore, they are not optimized for energy-constrained environments. In this regard, this study proposes an intelligent controller as a global allocation mechanism based on deep reinforcement learning (DRL), specifically on deep Q-network (DQN). The conceived mechanism optimizes energy consumption in SBCs by selecting the most suitable nodes across several clusters to deploy event requests in terms of nodes’ resources and events’ demands. A comparison with available allocation algorithms revealed that our solution required 28% fewer resource costs and reduced 35% the energy consumption in the clusters’ computing nodes while maintaining high levels of acceptance ratio.

Research and Application of the Obstacle Avoidance System for High-Speed Railway Tunnel Lining Inspection Train Based on Integrated 3D LiDAR and 2D Camera Machine Vision Technology

This study presents an innovative, intelligent obstacle avoidance module intended to significantly enhance the collision prevention capabilities of the robotic arm mechanism onboard a high-speed rail tunnel lining inspection train. The proposed module employs a fusion of ORB-SLAM3 and Normal Distribution Transform (NDT) point cloud registration techniques to achieve real-time point cloud densification, ensuring reliable detection of small-volume targets. By leveraging spatial filtering, cluster computation, and feature extraction, precise obstacle localization information is further obtained. A fusion of multi-modal data is achieved by jointly calibrating 3D LiDAR and camera images. Upon validation through field testing, it is demonstrated that the module can effectively detect obstacles with a minimum diameter of 0.5 cm, with an average deviation controlled within a 1–2 cm range and a safety margin of 3 cm, effectively preventing collisions. Compared to traditional obstacle avoidance sensors, this module provides information across more dimensions, offering robust support for the construction of powerful automated tunnel inspection control systems and digital twin lifecycle analysis techniques for railway tunnels.

Urban Mobility and Knowledge Extraction from Chaotic Time Series Data: A Comparative Analysis for Uncovering COVID-19 Effects

The COVID-19 pandemic exerted devastating effects on the global economy, public health, and urban life. The geographical consequences lie in spatial time series dimensions, but the chaotic characteristics hinder intuitive understanding. Aiming to discover the variance of urban mobility in the pandemic context, this article presents a hybrid clustering technique called whole time sequence mixture (WTSM). The combination of whole time and subsequence techniques ensures robustness to data volume, dimensionality, sampling, distortion tasks, and a prior constraint. For yellow taxi trips in New York City, the case study spanned the pre-, mid-, and postpandemic periods and compared the performance of the established methods (symbolic aggregate approximation and dynamic time warping) and WTSM techniques. Findings revealed COVID-19 trends and social restriction-induced mobility variations and determined that vaccine supply did not lead to immediate mobility restoration. Meanwhile, concepts of validation indexes and computational complexity corroborate the superiority of WTSM. The transient cluster arising from temporal dissimilarity is a unique finding of WTSM, which led to high cluster cohesion and separation. WTSM can obtain new knowledge and rationale for urgent government intervention and epidemic management, not sacrificing computational efficiency and clustering quality. With the improved storage capability and desire for multidimensional pattern extraction, the increased accessibility to the chaotic time series data sets could promote further studies and help administrative schemes.

Exploratory Data Analytics and PCA-Based Dimensionality Reduction for Improvement in Smart Meter Data Clustering

The smart meter sends the meter readings to the utilities at desired frequency allowing better visibility of consumer electricity consumption behaviour by providing more data points for in-depth analysis and for generating insights using advanced data analytics and data science techniques. The granulated data helps utilities in designing schemes for audience suitable for demand response management to shift the peak hour demand to off-peak hours. In this paper, a method is proposed for load profile segmentation which can be used by utilities for identifying the characteristics of different users and targeting those whose demand curve can be flattened during peak hours with various demand response management schemes. Firstly, exploratory data analysis is done on the cleaned dataset to find the optimal epoch size, understand the distribution of data in each epoch, and use it for dimensionality reduction. For reducing the clustering computation time, dimensionality reduction is done by around 64% using Principal Component Analysis. The first six principal components are identified as carrying maximum variance using the cumulative variance technique in each epoch. Unsupervised Machine Learning based k-means clustering technique is applied to these principal components. The optimal value of k is evaluated using the WCSS technique where k = 5 and k = 3 for residential and SME users respectively is found. The average silhouette coefficient for residential users is 0.48 and for SME users is 0.51. Hence, well-separated clusters are formed with minimum intra-cluster distance using PCA for dimensionality reduction which is used for load profile segmentation and Post Clustering Analysis.

Computational clustering reveals differentiated coronary artery calcium progression at prevalent levels of pulse wave velocity by classifying high-risk patients.

Many studies found that increased arterial stiffness is significantly associated with the presence and progression of Coronary Calcium Score (CCS). However, none so far have used machine learning algorithms to improve their value. Therefore, this study aims to evaluate the association between carotid-femoral Pulse Wave Velocity (cfPWV) and CCS score through computational clustering. We conducted a retrospective cross-sectional study using data from a cardiovascular risk screening program that included 377 participants. We used an unsupervised clustering algorithm using age, weight, height, blood pressure, heart rate, and cfPWV as input variables. Differences between cluster groups were analyzed through Chi-square and T-student tests. The association between (i) cfPWV and age groups, (ii) log (CCS) and age groups, and (iii) cfPWV and log(CCS) were addressed through linear regression analysis. Clusters were labeled post hoc based on cardiovascular risk. A "higher-risk group" had significantly higher left (0.76 vs. 0.70 mm, P < 0.001) and right (0.71 vs. 0.66 mm, P = 0.003) intima-media thickness, CCS (42 vs. 4 Agatston units, P = 0.012), and ascending (3.40 vs. 3.20 cm, P < 0.001) and descending (2.60 vs. 2.37 cm, P < 0.001) aorta diameters. Association with age appeared linear for cfPWV and exponential for log (CCS). The progression of the log (CCS) and cfPWV through age groups was steeper in the "higher-risk group" than in the "lower-risk group". cfPWV strongly correlated with CCS, and CCS progression over cfPWV differed among clusters. This finding could improve PWV as a "gate-keeper" of CCS testing and potentially enhance cardiovascular risk stratification.

Drone Swarm Path Planning for Mobile Edge Computing in Industrial Internet of Things

Drone-swarm-assisted mobile edge computing (MEC) provides extra computation and storage capacity for smart city applications and the industrial Internet of Things. To solve the problems of traditional fixed base stations in complex terrain, including cost of deployment, transmission loss of telecommunication, and limited coverage, this paper brings forward the unmanned aerial vehicles (UAVs) as mobile edge computing nodes in the air. For the purpose of matching the dynamic mobile devices and UAV trajectory, this paper raises a multi-UAVs-assisted mobile edge computing offloading algorithm based on global and local path planning controlled by ground station and onboard computer. Firstly, this paper considers a drone swarm scheduling and allocation strategy based on the priority of monitoring areas, UAVs residual energy and distance to target points, so that to minimize the global flight length and energy consumption. Secondly, based on user mobility, this paper calculates the optimal communication coverage of UAV, and jointly optimizes the local path planning and computing offloading, so that to maximize the number of offloading services and minimize the total latency in completing the computation task. Finally, based on the total latency and energy consumption of path planning and computation offloading, a UAV cluster computation offloading strategy with optimized energy efficiency is realized. Experimental results prove that the proposed algorithm can provide more offloading services while obtaining shorter path length and greater energy efficiency.

An optimization based framework for region wise optimal clusters in MR images using hybrid objective

Swarm Intelligence based methods are amongst the highly efficient approaches for optimization in image clustering. Optimal clustering has been studied in many real-world applications, such as medical and aerial image segmentation. Region wise clustering is a class of challenges in image region segmentation. Uncertain convergence and high computational load are critical issues in the region-wise image clustering due to local optimum and the NP-hard cluster computation. Meta-heuristics approaches are efficient to achieve global optimum by including better search space exploration techniques. This paper develops a framework for cluster optimization by selecting the seeds in pathological medical resonance (MR) images using a variant of firefly optimization. The heuristics based method uses Gaussian random walk for convergence that occasionally results in local optima; therefore, we have investigated the firefly method with more search space exploration techniques and improved region-wise objective. Our framework applies the levy flights for exploration and compared with other search spaces like Cauchy, and Gaussian random walk. The intra-cluster and inter-cluster-based hybrid objective is converged swiftly. The framework has been compared with two of its variants and three other meta-heuristic-based methods, namely simulated annealing, PSO, and Cuckoo search. The MSE, PSNR, structural similarity(SSIM), and feature similarity(FSIM) based evaluation indices are measured for normal and abnormal MR images and listed in table-5, 6. Reported indices values for our frame work are better than existing methods. Figure-9, 10, 11 compares the stochastic search spaces among Levy flights, Cauchy random walk, Gaussian random walk and observed Levy flights as better search space. In section-3.5, The convergence for proposed framework is shown for multi-objective function against the single objective in two normal and abnormal images. Single objective converged from 188 to 119 and multi-objective converged from 196 to 117 for first image. For second image, the single objective converged from 168 to 94 and multi-objective converged from 183 to 93. Finally, we have illustrated the convergence criteria and computation complexity on publicly available MR data sets.

K nearest neighbors classification of water masses in the western Alboran Sea using the sigma-pi diagram

Different classification techniques of water masses have been developped using the potential temperature-salinity (θ-S) diagram and its volumetric analysis. In this study, we propose a new method to automatically classify water masses via a supervised machine learning algorithm based on the K nearest neighbors (Knn), in the potential density and potential spicity (σ-π) coordinates. This method is applied to temperature and salinity data collected in the western side of the Alboran Sea during a glider mission, dedicated to sample the Western Alboran Gyre (WAG) in late winter 2021. The water masses in the studied region were classified into five different categories following a supervised learning process, based on ocean profile databases available on the region of interest. The results corroborate previous studies of the spatial distribution of water masses in the Alboran Sea, inferred from traditional method based on the expert analysis of the (θ-S) diagram, and suggest that this methodology is efficient and reliable for water masses classification. Compared to a classical clustering computation (herein k-means), this method is more appropriate in a region where the characteristics of the water masses change considerably in both space and time.

Simplifying Clustering with Graph Neural Networks

The objective functions used in spectral clustering are generally composed of two terms: i) a term that minimizes the local quadratic variation of the cluster assignments on the graph and; ii) a term that balances the clustering partition and helps avoiding degenerate solutions. This paper shows that a graph neural network, equipped with suitable message passing layers, can generate good cluster assignments by optimizing only a balancing term.Results on attributed graph datasets show the effectiveness of the proposed approach in terms of clustering performance and computation time.

R Libraries for Remote Sensing Data Classification by k-means Clustering and NDVI Computation in Congo River Basin, DRC

R Libraries for Remote Sensing Data Classification by k-means Clustering and NDVI Computation in Congo River Basin, DRC

R Libraries for Remote Sensing Data Classification by k-means Clustering and NDVI Computation in Congo River Basin, DRC

R Libraries for Remote Sensing Data Classification by k-means Clustering and NDVI Computation in Congo River Basin, DRC

R Libraries for Remote Sensing Data Classification by K-Means Clustering and NDVI Computation in Congo River Basin, DRC

In this paper, an image analysis framework is formulated for Landsat-8 Operational Land Imager and Thermal Infrared Sensor (OLI/TIRS) scenes using the R programming language. The libraries of R are shown to be effective in remote sensing data processing tasks, such as classification using k-means clustering and computing the Normalized Difference Vegetation Index (NDVI). The data are processed using an integration of the RStoolbox, terra, raster, rgdal and auxiliary packages of R. The proposed approach to image processing using R is designed to exploit the parameters of image bands as cues to detect land cover types and vegetation parameters corresponding to the spectral reflectance of the objects represented on the Earth’s surface. Our method is effective at processing the time series of the images taken at various periods to monitor the landscape dynamics in the middle part of the Congo River basin, Democratic Republic of the Congo (DRC). Whereas previous approaches primarily used Geographic Information System (GIS) software, we proposed to explicitly use the scripting methods for satellite image analysis by applying the extended functionality of R. The application of scripts for geospatial data is an effective and robust method compared with the traditional approaches due to its high automation and machine-based graphical processing. The algorithms of the R libraries are adjusted to spatial operations, such as projections and transformations, object topology, classification and map algebra. The data include Landsat-8 OLI-TIRS covering the three regions along the Congo river, Bumba, Basoko and Kisangani, for the years 2013, 2015 and 2022. We also validate the performance of graphical data handling for cartographic visualization using R libraries for visualising changes in land cover types by k-means clustering and calculation of the NDVI for vegetation analysis.

Federated Deep Reinforcement Learning-Based Caching and Bitrate Adaptation for VR Panoramic Video in Clustered MEC Networks

Virtual reality (VR) panoramic video is more expressive and experiential than traditional video. With the accelerated deployment of 5G networks, VR panoramic video has experienced explosive development. The large data volume and multi-viewport characteristics of VR panoramic videos make it more difficult to cache and transcode them in advance. Therefore, VR panoramic video services urgently need to provide powerful caching and computing power over the edge network. To address this problem, this paper establishes a hierarchical clustered mobile edge computing (MEC) network and develops a data perception-driven clustered-edge transmission model to meet the edge computing and caching capabilities required for VR panoramic video services. The joint optimization problem of caching and bitrate adaptation can be formulated as a Markov Decision Process (MDP). The federated deep reinforcement learning (FDRL) algorithm is proposed to solve the problem of caching and bitrate adaptation (called FDRL-CBA) for VR panoramic video services. The simulation results show that FDRL-CBA can outperform existing DRL-based methods in the same scenarios in terms of cache hit rate and quality of experience (QoE). In conclusion, this work developed a FDRL-CBA algorithm based on a data perception-driven clustered-edge transmission model, called Hierarchical Clustered MEC Networks. The proposed method can improve the performance of VR panoramic video services.

IOAM: A Novel Sensor Fusion-Based Wearable for Localization and Mapping

With the development of indoor location-based services (ILBS), the dual foot-mounted inertial navigation system (DF-INS) has been extensively used in many fields involving monitoring and direction-finding. It is a widespread ILBS implementation with considerable application potential in various areas such as firefighting and home care. However, the existing DF-INS is limited by a high inaccuracy rate due to the highly dynamic and non-stable stride length thresholds. The system also provides less clear and significant information visualization of a person’s position and the surrounding map. This study proposes a novel wearable-foot IOAM-inertial odometry and mapping to address the aforementioned issues. First, the person’s gait analysis is computed using the zero-velocity update (ZUPT) method with data fusion from ultrasound sensors placed on the inner side of the shoes. This study introduces a dynamic minimum centroid distance (MCD) algorithm to improve the existing extended Kalman filter (EKF) by limiting the stride length to a minimum range, significantly reducing the bias in data fusion. Then, a dual trajectory fusion (DTF) method is proposed to combine the left- and right-foot trajectories into a single center body of mass (CBoM) trajectory using ZUPT clustering and fusion weight computation. Next, ultrasound-type mapping is introduced to reconstruct the surrounding occupancy grid map (S-OGM) using the sphere projection method. The CBoM trajectory and S-OGM results were simultaneously visualized to provide comprehensive localization and mapping information. The results indicate a significant improvement with a lower root mean square error (RMSE = 1.2 m) than the existing methods.

Abstract 10397: Unbiased Cluster Modelling for Uncovering Natural Distribution of Doppler Derived Diastolic Variable Response to Exercise Among Patients Referred to Exercise Echocardiography

Introduction: Exercise-induced diastolic dysfunction is associated with poor prognosis. Hypothesis: We studied diastolic response to exercise based on the natural distribution of Doppler variables and its relationship to adverse outcomes. Methods: In a retrospective study, 813 patients referred for exercise echocardiography (ExE) for ischemia assessment were included (Age: 58±13 years, 57% females, EF:60±7%). A 2-step unsupervised cluster model was constructed using resting and post-exercise tissue Doppler early diastolic mitral annular velocity, post-exercise pulsed Doppler-derived mitral inflow velocity in early and late diastole, and post-exercise tricuspid regurgitation velocity. Clusters were compared for a composite outcome (death, cardiac rehospitalization, acute coronary syndrome, and need for future ischemia testing). Results: Exercise-induced ischemia occurred in 116(14%). The composite outcome occurred in 196(24%) patients over a median follow-up of 3.2 years. The model yielded 2 clusters labelled low and high risk for the composite outcome, Figure 1 A,B). Compared to the low risk cluster, patients in the high risk cluster were older, had more risk factors and comorbidities, worse exercise performance, worse resting and exercise diastolic variables, and more ischemic ExE. After stratification for ischemia, clusters resulted in significant separation of risk for the outcome, regardless of the presence or absence of ischemia (Figure 1 C,D). These findings persisted after multivariate adjustment. Conclusions: In patients referred to ExE for assessment of ischemia, diastolic response to exercise involves complex interactions among diastolic variables. Appreciation of such complexity seems to add prognostic value beyond ischemia, suggesting a need for novel approaches for characterization of the natural distribution of diastolic variables, such as computational unbiased clustering.

Tensor Train-Based Multiple Clusterings for Big Data in Cyber-Physical-Social Systems and Its Efficient Implementations

Multiple clusterings are conducive to discovering different data patterns hidden in data from different perspectives, so it has tremendous value in applications like community detection, resource recommendation, and gene expression, etc. To solve the problem that the existing multiple clustering approaches are mainly oriented to low-dimensional single-domain data and are not suitable for Big Data in Cyber-Physical-Social Systems (CPSS), a tensor-based multiple clustering (TMC) was proposed. However, as the scale of data continues to increase, data storage, computing load, and memory overhead will increase exponentially, leading to dimensional disasters and greatly affecting the efficiency of TMC. Therefore, a tensor train-based multiple clustering (TTMC) and its parallel computing method are studied in this paper. First, a tensor train (TT)-based multiple clustering parallel analytic and service framework is present. Then, a TT-based multi-linear attribute combination weight learning algorithm, a selective weighted tensor train distance, and the TTMC algorithm are put forward to improve the accuracy and efficiency of TMC. Furthermore, an efficient distributed parallel computing strategy of TTMC is designed by using TT core parallelism. Experimental results demonstrate that TTMC and its parallelization can significantly improve computation efficiency and clustering accuracy while reducing the running memory compared to the original TMC algorithm.

Perceptive Hierarchy of Facial Skin Lesions: An Eye-tracking Study.

Equal importance is given to every skin lesion in treatment guidelines and severity grading systems for facial lesions. Face recognition studies suggest differentially perceived areas of the human face. The aims of this study were to quantify the visual attention given to facial skin lesions and to explore their hierarchy. Eye movements were tracked in 118 participants who viewed 135 faces with facial skin lesions. The main effect of the image was significant (F[9, 1053]=15.631, p < 0.001, η2=0.118), which implied a difference in the total visual attention between images depicting skin lesions in different aesthetic units. Lesions in the frontal area received the highest area- specific attention, while lesions in the right parotid area had the smallest effect. Using objective computational clustering, 2 very distinct zones of visual attention were identified: the ocular, nasal, perioral, and frontal areas attracted high visual scrutiny and the remaining areas attracted little attention. However, the presence of skin lesions in the high-attention region resulted in a decrease in total visual attention to the face. The main effect of the aesthetic unit was significant (F[15, 1755]=202.178, p < 0.001, η2=0.633). More than 63% of attention-drawing potential of facial skin lesions depends on their anatomical location, which should be considered in disease severity grading and treatment planning.

Computational Clustering Applied to Mental Models for Understanding the Valley of Death in Innovation Processes

The Valley of Death is the gap between the completion of research and development (R&amp;D) projects and their transition to innovation. A key aspect to explain it are mindsets, which are one of the most complex to explain due to the number of factors they contain. What remains unclear is how people might have patterns of understanding the processes and activities that define mental models. This paper aims to explore how persons involved in R&amp;D activities have a pattern to understand the processes. Data for this study were collected using a survey applied to directives, coordinators, technology managers, intellectual property managers, researchers, and entrepreneurs in a group of 11 universities in Medellín (Colombia) through a computational clustering analysis. The main contribution of this article is the generation of five patterns or mental models, in which the different roles linked to R&amp;D converge, to this extent we could speak of shared mental models. One of the more significant findings that emerge from this study is that a simpler mental model with specific and relevant activities prioritised may work better than a complex one.

Binary black hole mergers from young massive clusters in the pair-instability supernova mass gap

Context. The recent discovery of the binary black hole (BBH) merger event GW190521, between two black holes (BHs) of ≈100 Msamp, in addition to other massive BBH merger events involving BHs within the pair-instability supernova (PSN) mass gap have sparked widespread debate on the origin of such extreme gravitational-wave (GW) events. GW190521 simultaneously triggers two critical questions: how BHs can appear within the ‘forbidden’ PSN gap and, if they do, how they get to participate in general-relativistic (GR) mergers. Aims. In this study, I investigate whether dynamical interactions in young massive clusters (YMCs) serve as a viable scenario for assembling PSN-gap BBH mergers. Methods. To that end, I explore a grid of 40 new evolutionary models of a representative YMC of initial mass and size Mcl = 7.5 × 104 Msamp (N ≈ 1.28 × 105) and rh = 2 pc, respectively. The model grid ranges over metallicity 0.0002 ≤ Z ≤ 0.02 and comprises initial cluster configurations of King central concentration parameters W0 = 7 and 9. In each model, all BH progenitor stars are initially in primordial binaries following observationally motivated distributions. All cluster models are evolved with the direct, relativistic N-body code NBODY7, incorporating up-to-date remnant formation, BH natal spin, and GR merger recoil schemes. Results. Binary black hole mergers from these model cluster computations agree well with the masses and effective spin parameters, χeff, of the events from the latest gravitational-wave transient catalogue (GWTC). In particular, GW190521-like, that is to say ≈200 Msamp, low χeff events are produced via a dynamical merger among BHs derived from star-star merger products. GW190403_051519-like, that is PSN-gap, highly asymmetric, high χeff events result from mergers involving BHs that are spun up via matter accretion or a binary interaction. The resulting present-day, differential intrinsic merger rate density, within the PSN gap, accommodates that from GWTC well. Conclusions. This study demonstrates that, subject to model uncertainties, the tandem of massive binary evolution and dynamical interactions in ≲100 Myr-old, low metallicity YMCs in the Universe can plausibly produce GR mergers involving PSN-gap BHs and in rates consistent with that from up-to-date GW observations. Such clusters can produce extreme events similar to GW190521 and GW190403_051519. The upper limit of the models’ GW190521-type event rate is within the corresponding LIGO-Virgo-KAGRA (LVK)-estimated rate limits, although the typical model rate lies below LVK’s lower limit. The present YMC models yield a merger rate density of 0−3.8 × 10−2 yr−1 Gpc−3 for GW190521-type events. They produce GW190403_051519-like events at a rate within 0−1.6 × 10−1 yr−1 Gpc−3 and their total BBH-merger yield within the PSN gap is 0−8.4 × 10−1 yr−1 Gpc−3.