Clustering Index Research Articles

Characterization of droplet clusters in a turbulent multiphase gas cloud using a model cough simulator

This study aims to investigate the transport of droplets ejected from an artificial cough simulator, which releases a turbulent puff of droplets into the surrounding air, closely resembling the human coughing process. The focus is on understanding droplet clustering within the multiphase gas cloud across various operating conditions that emulate the wide variation in the spray characteristics in actual human subjects owing to infection severity, age, and gender. Time-resolved particle image velocimetry (PIV) technique was employed to measure the velocity of both droplet and gas phases. It also facilitates the identification and characterization of droplet clusters through Voronoi analysis of the PIV images. The area and length scale of individual droplet clusters were measured, and the degree of droplet clustering was quantified using the clustering index and relative droplet number density within the clusters. Additionally, the interferometric laser imaging for droplet sizing technique was utilized for planar measurement of individual droplet sizes. The range of Stokes number indicated partial to poor response of the droplets to the turbulent eddies. The results reported, for the first time, the presence of droplet clusters in the simulated coughing process. The wide spectrum of cluster size and self-similar evolution of droplet clusters unveil a multi-scale clustering phenomenon, shedding light on the intricate dynamics of the respiratory droplet dispersion process. The study comprehensively investigates the role of injection pressure on droplet clustering and the spatial development of the clusters, revealing some interesting findings, which are discussed.

Exploring the Configurational Relationships between Urban Heat Island Patterns and the Built Environment: A Case Study of Beijing

The spatial heterogeneity of land surface temperature (LST) within cities is profoundly influenced by the built environment. Although significant progress has been made in the study of the urban thermal environment, there is still a lack of research on how the pattern and structural layout of the built environment affects the thermal environment. In this study, we take the Fifth Ring Road of Beijing as an example, invert the urban LST on the basis of multisource spatial data, characterize the built environment, and use k-means cluster analysis to investigate the main influencing factors of the LST in different functional areas and building patterns within the city, as well as the spatial relationship between the built environment and the urban LST. The results show the following: (1) The urban heat island (UHI) effect occurs to varying degrees over a large part of the study area, and these UHI areas are mainly concentrated in the southwestern part of the city, forming a large contiguous area between the second and fifth ring roads. (2) Class 1 is dominated by transport blocks, Class 3 is dominated by commercial blocks, and Class 5 is dominated by green space blocks, with a clustering index of 0.38. (3) The high-density, high-height class (HH-Class 2) has a greater number of blocks distributed in a ring shape around the periphery of the second ring road. The high-density, low-height class (HL-Class 2) has a relatively small number of blocks but a relatively large area, and the largest blocks are located in the western part of the study area. (4) In the HH and HL building patterns, extreme heat scenarios often occur; from the perspective of functional areas, the probability of extreme heat in the transport block is much higher than that of other functional areas, and except for the HH scenario, the green space functional area plays a very important role in reducing the temperature. This study explores the characteristics of the built environment that influence the urban LST from the perspective of different urban functional zones in cities to provide decision support for quantitative territorial spatial planning, optimization, and management.

Spatial Pattern of Host Tree Size, Rather than of Host Tree Itself, Affects the Infection Likelihood of a Fungal Stem Disease.

The spatial pattern of diseased forest trees is a product of the spatial pattern of host trees and the disease itself. Previous studies have focused on describing the spatial pattern of diseased host trees, and it remains largely unknown whether an antecedent spatial pattern of host/nonhost trees affects the infection pattern of a disease and how large the effect sizes of the spatial pattern of host/nonhost trees and host size are. The results from trivariate random labeling showed that the antecedent pattern of the host ash tree, Fraxinus mandshurica, but not of nonhost tree species, impacted the infection pattern of a stem fungal disease caused by Inonotus hispidus. To investigate the effect size of the spatial pattern of ash trees, we employed the SADIE (Spatial Analysis by Distance IndicEs) aggregation index and clustering index as predictors in the GLMs. Globally, the spatial pattern (vi index) of ash trees did not affect the infection likelihood of the focal tree; however, the spatial pattern of DBH (diameter at breast height) of ash trees significantly affected the infection likelihood of the focal tree. We sampled a series of circular plots with different radii to investigate the spatial pattern effect of host size on the infection likelihood of the focal tree locally. The results showed that the location (patch/gap) of the DBH of the focal tree, rather than that of the focal tree itself, significantly affected its infection likelihood in most plots of the investigated sizes. A meta-analysis was employed to settle the discrepancy between plots of different sizes, which led to results consistent with those of global studies. The results from meta-regression showed that plot size had no significant effects.

Utility of the Clustering Index method for diagnosing neuromuscular disorders as compared with needle electromyography.

Concentric needle electromyography (CNEMG) is an essential examination for evaluating neuromuscular disorders, although pain is a drawback. Clustering Index (CI) method is a non-invasive quantitative analysis for surface electromyography (SEMG) that evaluates whether the signal area is clustered into the few large motor unit potentials (MUPs) or is evenly distributed. However, the diagnostic yield of the CI method in comparison with CNEMG is not known. In this study, we aimed to compare the sensitivity of the CI method with MUP parameters in CNEMG for diagnosing neurogenic or myogenic disorders. We retrospectively identified patients for whom both SEMG and CNEMG were performed on the same tibialis anterior (TA) muscle. In CNEMG, seven MUP parameters were evaluated, including size index (SI) and revised size indices for neurogenic (rSIn) and myogenic (rSIm) disorders. Identified were 21 patients with neurogenic and 21 patients with myogenic disorders. Control data were constructed from 30 control subjects. The sensitivities of the CI method for the neurogenic and myogenic groups were 76% and 62%, respectively, which were not significantly different from MUP parameters, except for being significantly higher than those of amplitude and duration for myopathy (24%). Among MUP parameters, the sensitivities of rSIn (62%) and rSIm (57%) for myopathy were significantly higher than those of amplitude and duration. The CI method significantly correlated with the strength of the TA muscle in myopathy. The CI method, having comparable diagnostic yields to MUP parameters, is promising as a non-invasive diagnostic measure.

Landsat images and GIS techniques as key tools for historical analysis of landscape change and fragmentation

Monitoring and evaluation of landscape fragmentation is important in numerous research areas, such as natural resource protection and management, sustainable development, and climate change. One of the main challenges in image classification is the intricate selection of parameters, as the optimal combination significantly affects the accuracy and reliability of the final results. This research aimed to analyze landscape change and fragmentation in northwestern Peru. We utilized accurate land cover and land use (LULC) maps derived from Landsat imagery using Google Earth Engine (GEE) and ArcGIS software. For this, we identified the best dataset based on its highest overall accuracy, and kappa index; then we performed an analysis of variance (ANOVA) to assess the differences in accuracies among the datasets, finally, we obtained the LULC and fragmentation maps and analyzed them. We generated 31 datasets resulting from the combination of spectral bands, indices of vegetation, water, soil and clusters. Our analysis revealed that dataset 19, incorporating spectral bands along with water and soil indices, emerged as the optimal choice. Regarding the number of trees utilized in classification, we determined that using between 10 and 400 decision trees in Random Forest classification doesn't significantly affect overall accuracy or the Kappa index, but we observed a slight cumulative increase in accuracy metrics when using 100 decision trees. Additionally, between 1989 and 2023, the categories Artificial surfaces, Agricultural areas, and Scrub/ Herbaceous vegetation exhibit a positive rate of change, while the categories Forest and Open spaces with little or no vegetation display a decreasing trend. Consequently, the areas of patches and perforated have expanded in terms of area units, contributing to a reduction in forested areas (Core 3) due to fragmentation. As a result, forested areas smaller than 500 acres (Core 1 and 2) have increased. Finally, our research provides a methodological framework for image classification and assessment of landscape change and fragmentation, crucial information for decision makers in a current agricultural zone of northwestern Peru.

Clustering of high-dimensional observations

We present a novel clustering method for high-dimensional, low sample size (HDLSS) data. The method is distance-based, takes advantage of the distance concentration phenomenon and the limiting values of the dissimilarity indices to construct clusters. We describe an algorithm that orders each row of the dissimilarity matrix to estimate the change points, which define cluster boundaries. We construct an agreement matrix of the Rand indices of the row clusters. The minimum of the row sum of the agreement matrix provides us with the best clusters. We prove that the new method achieves perfect clustering as the number of features diverges for a fixed sample size. Several examples are presented to illustrate the proposed method. We compare the new method with four other clustering techniques, including high-dimensional k-means, minimal spanning tree and Hierarchical Scan. The clustering methods are applied to the Lymphoma data set.

A multi-view ensemble clustering approach using joint entropy

Multi-view clustering can capture complementary and consistent information from different views, which is a core research topic in the fields of machine learning and pattern recognition. However, existing multi-view clustering methods ignore the quality of each base clustering cluster, which may affect the clustering performance. In this study, a multi-view ensemble clustering approach is proposed using joint entropy to evaluate the base clustering clusters. Firstly, an uncertainty index of base clustering clusters is defined using joint entropy which characterizes the importance and quality of each cluster. Secondly, a weighted co-association matrix is constructed using the uncertainty index, and useless entries are removed from the matrix, making the co-association matrix more reasonable. Thirdly, a candidate clusters selection strategy based on the stability index and a Multi-view Ensemble Clustering (MvEC-DoS) algorithm is proposed. In the end, experiments on five benchmark datasets validate the efficacy of our approach compared to other state-of-the-art methods.

3D clustering of gene expression data from systemic autoinflammatory diseases using self-organizing maps (Clust3D)

Background and objectiveSystemic autoinflammatory diseases (SAIDs) are characterized by widespread inflammation, but for most of them there is a lack of specific biomarkers for accurate diagnosis. Although a number of machine learning algorithms have been used to analyze SAID datasets, aiding in the discovery of novel biomarkers, there is a growing recognition of the importance of SAID timeseries clustering, as it can capture the temporal dynamics of gene expression patterns. MethodologyThis paper proposes a novel clustering methodology to efficiently associate three-dimensional data. The algorithm utilizes competitive learning to create a self-organizing neural network and adjust neuron positions in time-dependent and high dimensional feature space in order to assign them as clustering centers. The quantitative evaluation of the clustering was based on well-known clustering indices. Furthermore, a differential expression analysis and classification pipeline was employed to assess the capability of the proposed methodology to extract more accurate pathway-specific genes from its clusters. For that, a comparative analysis was also conducted against a heuristic timeseries clustering method. ResultsThe proposed methodology achieved better overall clustering indices scores and classification metrics using genes derived from its clusters. Notable cases include a threefold increase in the Calinski-Harabasz clustering index, a twofold improvement in the Davies–Bouldin clustering index and a ∼60% increase in the classification specificity score. ConclusionA novel clustering methodology was developed and applied on several gene expression timeseries datasets from systemic autoinflammatory diseases, and its ability to efficiently produce well separated clusters compared to existing heuristic methods was demonstrated.

New insights on clusters stacking mechanism within medium range in Fe-Al amorphous alloys

Molecular dynamics simulation has been employed to investigate the stacking mechanism of short-range-order structures in Fe-Al amorphous alloys. We explore the effect of Al content (0–50 %) on the nearest-neighbour correlation index and stacking degree between their main clusters. The results show that the correlation index changes little over the Al content range, and the stacking degree decreases with the Al content increasing on the whole. Through the spatial correlation analysis, it is found that the clusters with similar structural symmetry are more likely to connect. Moreover, we find that the stacking degree depends on the number of clusters and their correlation index, i.e., the stacking degree is found to increase as the number or correlation index of clusters increases, but the cluster number shows a greater influence on the stacking degree compared to their correlation index. We believe that these are the inherent stacking features in medium range in Fe-Al amorphous alloys, which can also be extended to other amorphous alloys. This work provides a deeper insight into the spatial relationship of short-range-order structures and the stacking of clusters from short to medium range.

Security Improvement of UAV Group IoT in Polygonal Area Based on Similarity Clustering Algorithm

Abstract Currently, unmanned aerial vehicle (UAV) group Internet of Things (IoT) has poor stability, poor endurance, and low security. In this regard, we report a polygon area UAV group IoT clustering algorithm for large-scale, high-speed mobile IoT environment. First, the UAV group IoT was divided into a polygon area and a few cluster heads were used to cover the polygon area uniformly. Then, the clustering index in the maximum speed similarity clustering algorithm was induced into the weighted clustering algorithm. Simultaneously, the link retention rate, relative node degree, and node residual energy ratio were improved to select the IoT node with the maximum weight as the cluster head. Finally, the IoT security was improved by reputation calculation. The results show that the clustering algorithm in this paper can reduce the number of clusters and lower the handover rate between clusters, improve the stability of clustering, prolong the survival time of the minimum node, improve the overall endurance of the IoT, and have high security.

DEA-based internal validity index for clustering

Internal validity indices are crucial in evaluating the quality of clustering results, serving as valuable tools for comparing various clustering algorithms and determining the optimal number of clusters for datasets. Most existing internal validity indices use the worst-case scenario to represent the overall validity. Moreover, some indices assign equal weights to distances among different clusters, even when these distances might have varying degrees of influence on the overall validity. Data envelopment analysis (DEA) is an effective technique for evaluating the performance of decision-making units through the computation of the ratio of the weighted sum of outputs to the weighted sum of inputs. The weight assigned to each indicator signifies its degree of influence on efficiency. Furthermore, DEA can be viewed as a multiple-criteria evaluation methodology, wherein inputs and outputs are two sets of performance criteria. We propose a DEA-based internal validity index (DEAI) to evaluate the validity of the clustering results. In this approach, intra-cluster compactness and inter-cluster separation are employed for determining the input(s) and output(s). The DEAI is then applied to the artificial datasets and empirical examples. Experimental results illustrate that DEAI outperforms six classic internal validity indices in accurately identifying the optimal cluster across all 10 datasets.

Clustering index analysis on EMG-Torque relation-based representation of complex neuromuscular changes after spinal cord injury

Spinal cord injury (SCI) resulting in complex neuromuscular pathology is not sufficiently well understood. To better quantify neuromuscular changes after SCI, this study uses a clustering index (CI) method for surface electromyography (sEMG) clustering representation to investigate the relation between sEMG and torque in SCI survivors. The sEMG signals were recorded from 13 subjects with SCI and 13 gender-age matched able-bodied subjects during isometric contraction of the biceps brachii muscle at different torque levels using a linear electrode array. Two torque representations, maximum voluntary contraction (MVC%) and absolute torque, were used. CI values were calculated for sEMG. Regression analyses were performed on CI values and torque levels of elbow flexion, revealing a strong linear relationship. The slopes of regressions between SCI survivors and control subjects were compared. The findings indicated that the range of distribution of CI values and slopes was greater in subjects with SCI than in control subjects (p < 0.05). The increase or decrease in slope was also observed at the individual level. This suggests that the CI and its sEMG clustering-torque relation may serve as valuable quantitative indicators for determining neuromuscular lesions after SCI, contributing to the development of effective rehabilitation strategies for improving motor performance.

Limit-hitting exciting effects: Modeling jump dependencies in stock markets adhering to daily price-limit rules

Price limits are widely implemented in stock markets worldwide; however, they are rarely considered in financial models. In this study, we propose a model specifically designed for asset prices that adhere to daily price-limit mechanisms. Our model captures the interdependence among limit-hitting events and other small price jumps by using a multivariate mutually-exciting point process. It is applicable to any stock market with a multi-layer price limit mechanism. By analyzing data from all publicly listed A-share stocks in China from 2007 to 2021, we demonstrate that our model outperforms other classic models in terms of goodness of fit. Additionally, we find that limit-hitting jumps, as opposed to inconspicuous small price jumps, have a higher propensity to attract investors' attention and result in subsequent price jumps. We further construct a clustering index based on the model parameters and investigate its determinants.

Optimal Coordination Of Multiple Setting Group Type Overcurrent Relays

Relay coordination optimization issues are multi-objective, multi-constrained problems that aim to reduce each relay's operation time while satisfying all relevant constraints. Fulfillment of the coordination time interval (CTI) and time limitations for relay operations are the coordination constraints. To solve coordination optimization issues, decision-making variables are defined such that the objective function is minimized with due satisfaction of necessary constraints. The majority of the most recent research that is currently available defines relay coordination for fixed network topologies, but in practice, it varies for a variety of reasons, including maintenance and element failures. Over current relays now have the capability of storing multiple relay settings but the number of relay settings that can be stored in relays is far less than the possible number of network configurations, which is why a suitable clustering technique such as K-Means can be used to cluster out various network topologies using a proper clustering index. Because both the objective function and the constraints are time dependent, time dependent attributes such as average relay operating time are used as the clustering index instead of the conventional index fault current deviation.

Graphical, statistical and index-based techniques integrated for identifying the hydrochemical fingerprints and groundwater quality of In Salah, Algerian Sahara.

Groundwater, a predominant reservoir of freshwater, plays a critical role in providing a sustainable potable water and water for agricultural and industry uses in the In Salah desert region of Algeria. This research collected 82 underground water samples from Albian aquifers to assess water quality and identify hydrogeochemical processes influencing mineralization. To achieve this objective, various methods were employed to evaluate water quality based on its intended uses. The drinking water quality index utilized revealed the water potability status, while the indicators of irrigation potability were employed to evaluate its quality for agricultural purposes. Additionally, an assessment of groundwater susceptibility to corrosion and scaling in an industrial context was conducted using several indices, e.g., Langelier index, Larson-Skold index, Ryznar index, chloride-sulfate mass ratio, Puckorius index, aggressiveness index, and the Revelle index. The findings of this study revealed thatthe groundwater quality for consumption fell into four categories: good (2.44%), fair (29.27%), poor (65.85%), and non-potable (2.44%). Concerning agricultural irrigation, the indexical results indicated that 15.85% of the waters exhibited adequate quality, while 84.15% were questionable for irrigation. Calculations based on various corrosion and scaling evaluation indices showed that most wells were prone to corrosion, with a tendency for calcium bicarbonate deposit formation. Furthermore, the hydrochemical study identified three water types: Na-Cl (53.66%), Ca-Mg-Cl (37.80%), and Ca-Cl (8.54%) waters. Analyses of correlation matrices, R-type clustering, factor loadings, Gibbs diagrams, scatterplots, and chloro-alkaline indices highlighted that the chemistry of the Albian groundwater is fundamentally impacted by a number of processes such as silicate weathering, evaporite dissolution, ionic exchange, and anthropogenic inputs, that played impactful role in the aquifer's water chemistry.

Data-driven determination of zooplankton bioregions and robustness analysis

Identifying biogeographic regions through cluster analysis of species distribution data is a common method for partitioning ecosystems. Selecting the appropriate cluster analysis method requires a comparison of multiple algorithms. In this study, we demonstrate a data-driven process to select a method for bioregionalization based on community data and test its robustness to data variability following these steps:•We aggregated and curated zooplankton community observations from expeditions in the Northeast Pacific.•We determined the best bioregionalization approach by comparing nine cluster analysis methods using ten goodness of clustering indices.•We evaluated the robustness of the bioregionalization to different sources of sampling and taxonomic variability by comparing the bioregionalization of the overall dataset with bioregionalizations of subsets of the data.The K-means clustering of the log-chord transformed abundance was selected as the optimal method for bioregionalization of the zooplankton dataset. This clustering resulted in the emergence of four bioregions along the cross-shelf gradient: the Offshore, Deep Shelf, Nearshore, and Deep Fjord bioregions. The robustness analyses demonstrated that the bioregionalization was consistent despite variability in the spatial and temporal frequency of sampling, sampling methodology, and taxonomic coverage.

The Impact of Multimorbidity Patterns on Changes in Physical Activity and Physical Capacity Among Older Adults Participating in a Year-Long Exercise Intervention.

This study investigated the impact of multimorbidity patterns on physical activity and capacity outcomes over the course of a year-long exercise intervention, and on physical activity 1year later. Participants were 314 physically inactive community-dwelling men and women aged 70-85years, with no contraindications for exercise at baseline. Physical activity was self-reported. Physical capacity measurements included five-time chair-stand time, 6-minute walking distance, and maximal isometric knee-extension strength. The intervention included supervised and home-based strength, balance, and walking exercises. Multimorbidity patterns comprised physician-diagnosed chronic disease conditions as a predictor cluster and body mass index as a measure of obesity. Multimorbidity patterns explained 0%-12% of baseline variance and 0%-3% of the change in outcomes. The magnitude and direction of the impact of unique conditions varied by outcome, time point, and sex. Multimorbid older adults with no contraindications for exercise may benefit from multimodal physical training.

Regional Variation in Cardiovascular Genes Enables a Tractable Genome Editing Strategy.

To realize the potential of genome engineering therapeutics, tractable strategies must be identified that balance personalized therapy with the need for off-the-shelf availability. We hypothesized that regional clustering of pathogenic variants can inform the design of rational prime editing therapeutics to treat the majority of genetic cardiovascular diseases with a limited number of reagents. We collated 2435 high-confidence pathogenic/likely pathogenic (P/LP) variants in 82 cardiovascular disease genes from ClinVar. We assessed the regional density of these variants by defining a regional clustering index. We then combined a highly active base editor with prime editing to demonstrate the feasibility of a P/LP hotspot-directed genome engineering therapeutic strategy in vitro. P/LP variants in cardiovascular disease genes display higher regional density than rare variants found in the general population. P/LP missense variants displayed higher average regional density than P/LP truncating variants. Following hypermutagenesis at a pathogenic hotspot, mean prime editing efficiency across introduced variants was 57±27%. Designing therapeutics that target pathogenic hotspots will not only address known missense P/LP variants but also novel P/LP variants identified in these hotspots as well. Moreover, the clustering of P/LP missense rather than truncating variants in these hotspots suggests that prime editing technology is particularly valuable for dominant negative disease. Although prime editing technology in relation to cardiac health continues to improve, this study presents an approach to targeting the most impactful regions of the genome for inherited cardiovascular disease.

T-Distributed Stochastic Neighbor Embedding for Co-Representation Learning

Co-clustering is the simultaneous clustering of the samples and attributes of a data matrix that provides deeper insight into data than traditional clustering. However, there is a lack of representation learning algorithms that serve this mechanism of co-clustering, and the current representation learning algorithms are limited to the sample perspective and lack the use of information in the attribute perspective. To solve this problem, in this article, ctSNE , a co-representation learning model based on t-distributed stochastic neighbor embedding, is proposed for unsupervised co-clustering, where ctSNE makes the dataset representation outputted more discriminative of row and column clusters (i.e. co-discrimination). On the basis of t-distributed stochastic neighbor embedding retaining the sample data distribution and local data structure, the philosophy of collaboration is introduced (i.e., row and column hidden relationship information) so that the ctSNE model is equipped with co-representation learning capability, which can effectively improve the performance of co-clustering. To prove the effectiveness of the ctSNE model, several classic co-clustering algorithms are used to check the co-representation performance of ctSNE, and a novel internal index based on an internal clustering index, known as total inertia, is proposed to demonstrate the effect of co-clustering. The numerous experimental results show that ctSNE has tremendous co-representation capability and can significantly improve the performance of co-clustering algorithms.

Optimization of medium components for protein production by Escherichia coli with a high-throughput pipeline that uses a deep neural network

To optimize rapidly the medium for green fluorescent protein expression by Escherichia coli with an introduced plasmid, pRSET/emGFP, a single-cycle optimization pipeline was applied. The pipeline included a deep neural network (DNN) and mathematical optimization algorithms with simultaneous optimization of 18 medium components. To evaluate the DNN data sampling method, two methods, orthogonal array (OA) and Latin hypercube sampling (LHS), were used to design 64 initial media for each sampling method. The OA- and LHS-based data sampling resulted in green fluorescent protein fluorescence intensities of 0.088× 103-1.85× 104 and 3.30× 103-1.50× 104, respectively. Fifty DNN models were built using the OA and LHS datasets. Hold-out validation was performed using 15% test of OA and LHS data. Mean square errors of the DNN models were 0.015-0.64, indicating the estimation accuracies were sufficient. However, the sensitivities of components in the DNN models varied and were grouped into six major classes by the index of k-means clustering. A representative model was selected for each class. Mathematical optimization algorithms using Bayesian optimization and genetic algorithm were applied to the representative models, and representative optimized medium (OM) compositions were selected by k-means clustering from the proposed OMs. A total of 54 OMs were obtained from the OA and LHS datasets. In the validating cultivation, the best OMs of OA and LHS were 2.12-fold and 2.13-fold higher, respectively, than those of the learning data.