Search

The growing complexity of biological data has spurred the development of innovative computational techniques to extract meaningful information and uncover hidden patterns within vast datasets. Biological networks, such as gene regulatory networks and protein-protein interaction networks, hold critical insights into biological features' connections and functions. Integrating and analyzing high-dimensional data, particularly in gene expression studies, stands prominent among the challenges in deciphering these networks. Clustering methods play a crucial role in addressing these challenges, with spectral clustering emerging as a potent unsupervised technique considering intrinsic geometric structures. However, spectral clustering's user-defined cluster number can lead to inconsistent and sometimes orthogonal clustering regimes. We propose the Multi-layer Bundling (MLB) method to address this limitation, combining multiple prominent clustering regimes to offer a comprehensive data view. We call the outcome clusters "bundles". This approach refines clustering outcomes, unravels hierarchical organization, and identifies bridge elements mediating communication between network components. By layering clustering results, MLB provides a global-to-local view of biological feature clusters enabling insights into intricate biological systems. Furthermore, the method enhances bundle network predictions by integrating the bundle co-cluster matrix with the affinity matrix. The versatility of MLB extends beyond biological networks, making it applicable to various domains where understanding complex relationships and patterns is needed. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of fitting concentric ellipses under general assumptions. We study two methods of obtaining an estimator of the concentric ellipse parameters under this model, namely the least squares and the gradient weighted algebraic fits. We address some practical issues in obtaining these estimators. In this paper, we study the statistical properties of those estimators and we developed a refinement with the highest accuracy for each estimator. We also address a practical issue in concentric ellipse fitting, namely, that each observation in the data set should be recognized as belonging to only one of the concentric ellipses. Most well-known clustering methods, such as spectral clustering, fail for this problem. We propose a clustering approach that can effectively be used for the implementation of our model. Our theory has been validated through intensive numerical experiments on synthetic and real data. [ABSTRACT FROM AUTHOR]

This study offers a typology of the Social Innovation (SI) field. A sample of 5,152 documents from the Scopus database is screened using a clustering process. Using measures for inferring similarities from co-occurrence data, we identified the main themes addressed in the field, presenting a collection of keywords structured into clusters and categories. Bibliographic coupling and co-citation clustering find significant related documents and the VOS algorithm maps two sets of relevant clusters. The maps of these bibliometric analyses were contrasted through an analytical process identifying and classifying the research questions in each. This combined methodology proposes a typology that contributes to the integration of the different disciplines that structure the field by bridging the boundaries of the knowledge areas generated during the conceptualisation process. The typology confirms a matrix of six consolidated disciplines (transformation, social entrepreneurship, design, theory, territorial development, and governance) as well as three emerging clusters, a human cluster related to demographic challenges and two central sets bringing together specific aspects of social technologies and Sustainable Development Goals (SDGs). This classification provides a picture of the current intellectual structure of the field, pulling together publications relevant to the different academic communities and providing clarity and a more manageable approach to handling the large volume of information faced when designing and substantiating systematic studies and future research on SI. [ABSTRACT FROM AUTHOR]

In the field of hot rolling big data, the presence of different steel types, specifications, and data heterogeneity poses significant challenges to the accuracy and stability of using single machine learning regression technology for prediction. Therefore, this study proposes a hot‐rolled strip crown prediction method that combines data clustering and fusion modeling. First, this article introduces a relevant mechanism for designing cluster strategies. The optimal clustering strategy is determined through comparative experiments using rolling process parameters, strip size, and main material components as the clustering features. Subsequently, the K‐Means++ algorithm is used to effectively cluster the training and testing datasets based on this strategy, generating multiple clusters for both datasets. Finally, this study establishes seven different training models to match the most suitable regression prediction model for each cluster, and matching between each cluster and the model is determined through rigorous testing. The evaluation of the fusion model shows an R2 value of 0.829 and a root mean square error value of 3.974. The experimental results show that the proposed method outperforms traditional methods in solving the challenges of multiclass classification and data heterogeneity, providing strong data support for the intelligent control of the hot‐rolled strip crown in the future. [ABSTRACT FROM AUTHOR]

By analyzing the failure mechanisms, crashworthiness characteristics of FW composite tubes subjected to two modes of progressive damage and catastrophic failure are investigated using acoustic emission technique and numerical method. The AE signals of ±45° composite tubes were classified using hierarchical and wavelet transform methods, and based on the realistic and three-dimensional geometrical architecture of tubular structures, the microstructural finite element model was developed using Catia and ABAQUS software. Then deformation patterns and the impression of each mechanism on the crashworthiness characteristics were assessed. Results indicated that fiber breakage and fiber/matrix debonding could likely control the higher percentage of damage. By changing the type of modes from progressive damage to catastrophic failure, the percentage of matrix cracking increases, the fiber/matrix separation decreases, and the failure behavior become dominated by local buckling. Comparing the FE simulation with experimental results, we found the proposed 3D model can reasonably predict the pre-crushing, post-crushing, and material densification. [ABSTRACT FROM AUTHOR]

The lack of discharge observations and reliable drainage information is a pervasive problem in urban catchments, resulting in difficulties in parameterizing urban hydrological models. Current parameterization methods for ungauged urban catchments mostly rely on subjective experiences or simplified models, resulting in inadequate accuracy for urban flood prediction. Parameter regionalization has been widely used to tackle model parameterization issues, but has rarely been employed for urban hydrological models. How to conduct effective parameter regionalization for urban hydrological models remains to be investigated. Here we propose a parameter regionalization framework (PRF) that integrates donor catchment clustering and the optimal regression‐based methods in each cluster. The PRF is applied to an urban hydrological model, the Time Variant Gain Model in urban areas (TVGM_Urban), in 37 urban catchments in Shenzhen City, China. We first show satisfactory flood simulation performance of TVGM_Urban for all urban catchments. Subsequently, we employ the PRF for parameter regionalization of TVGM_Urban. PRF classifies 37 urban catchments into three groups, and the partial least‐squares regression is identified as optimal regression‐based method for Groups 1 and 2, while the random forest model is found to be best for Group 3. To evaluate the simulation performance of PRF, we compare it with eight single regionalization methods. The results indicate better simulation performance and lower uncertainty of PRF, and donor catchment clustering can effectively enhance the simulation performance of linear regression‐based methods. Lastly, we identify curve number, land cover area ratios, and slope as critical factors for most TVGM_Urban parameters based on PRF results. Plain Language Summary: Frequent urban flooding and waterlogging pose significant threats to human life and property. Accurate modeling urban floods is crucial for minimizing damages caused by urban flooding. However, the delayed initiation of urban observations has resulted in prevalent data scarcity in urban areas, leading to difficulties and low accuracy in estimating model parameters for ungauged urban catchments. This study proposes a parameter regionalization framework for an urban hydrological model to improve the accuracy of flood simulations in ungauged urban catchments. This framework involves classifying gauged catchments into multiple groups and identifying the optimal regression‐based methods in each group for parameter regionalization. The flood simulation performance and stability of the parameter regionalization framework is evaluated by comparing it with eight widely‐used regionalization methods in 37 urban catchments located in Shenzhen City, China. Our results show that the classification of gauged catchments can significantly improve the simulation performance of linear regression methods. Additionally, the parameter regionalization framework outperforms the other regionalization methods for both urban flood simulation in the study area and urban flood prediction in the ungauged urban catchments. This study provides a novel and efficient parameterization method for accurate flood prediction in ungauged urban catchments to reduce potential losses. Key Points: TVGM_Urban presents satisfying performance in urban flood modelingThe PRF integrating donor catchment clustering and optimal regression methods in each cluster outperforms single regionalization methodsCurve number, land cover area ratio, and slope are critical factors for runoff generation parameters of TVGM_Urban [ABSTRACT FROM AUTHOR]

Distributed Denial of Service (DDoS) attacks have increased in frequency and sophistication over the last ten years. Part of the challenge of defending against such attacks requires the analysis of very large volumes of data. Metaheuristic algorithms can assist in selecting relevant features from the network traffic data for use in DDoS detection models. By efficiently exploring different combinations of features, these methods can identify subsets that are informative for distinguishing between normal and attack traffic. However, identifying an optimized solution in this area is an open research question. Tuning the parameters of metaheuristic search techniques in the optimization process is critical. In this study, a switching approximation is used in a variety of metaheuristic search techniques. This approximation is used to find the best solution for the analysis of the network traffic features in either lower or upper values between 0 and 1. We compare the fine-tuning of this parameter against standard approaches and find that it is not substantially better than the BestFirst algorithm (a standard default approach for feature selection). This study contributes to the literature by testing and eliminating various fine-tuning strategies for the metaheuristic approach. [ABSTRACT FROM AUTHOR]

The vitality of small and medium enterprises (SMEs) is integral to the economic fortification of nations, necessitating refined enhancement mechanisms from governmental bodies. Distinctive in their developmental trajectory, SMEs present a unique challenge in prioritizing interventions. A Decision Support System (DSS), employing a hybridized methodology of Clustering and Analytic Hierarchy Process-Technique for Order Preference by Similarity to Ideal Solution (AHP-TOPSIS), is proposed to facilitate the stratification of SMEs and guide governmental action based on a hierarchical scale of priorities. In this study, K-Means clustering was adopted for the segmentation of SMEs, leveraging its capability to efficiently partition high-dimensional data with minimal error. Subsequently, the TOPSIS method was utilized to rank SMEs within each cluster. However, the critical step of computing criteria weights to ascertain their relative importance was achieved through the AHP method. The latter effectively addresses multivariate considerations encompassing both quantitative and qualitative criteria through pairwise comparison matrices. The research incorporated 11 attributes, encompassing essential characteristics intrinsic to business entities, such as business name, location, operational status, sector, tax identification number, workforce size, average revenue, production costs, operational challenges, credit accessibility, and external financing needs. The clustering process, executed via Self-Organizing Maps (SOM), yielded optimal clusters at 1000 epochs, evidenced by a Davies Bouldin Index (DBI) of 0.74785, translating to an accuracy of 91.2601%. Notably, the SOM's performance in clustering SME data surpassed that of KMeans, as demonstrated by superior results in both Sum of Squared Error (SSE) and DBI metrics, thus showcasing its proficiency in managing data with heterogeneous criteria. The methodology engenders an n-cluster output, from which members are earmarked for priority ranking. AHP-derived weights are calculated, with a Consistency Ratio (CR) exceeding 0 denoting consistency, thereby determining the significance of each SME. TOPSIS calculations then ascribe the final score, delineating the SMEs' standings. This integrated DSS framework presents a robust tool for policymakers, ensuring targeted and efficient allocation of resources towards the advancement of SMEs. [ABSTRACT FROM AUTHOR]

In order to improve the ability of power transmission system to cope with compound faults on the communication side and power side, a cyber-physical collaborative restoration strategy is proposed. First, according to the information system's role in fault diagnosis, remote control of equipment maintenance and automatic output adjustment of generator restoration, a cyber-physical coupling model is proposed. On this basis, a collaborative restoration model of power transmission system is established by studying interactions among maintenance schedule paths, information system operation, and power system operation. Based on power flow linearization and the large M-∊ method, the above model is transformed into a mixed integer linear programming model, whose computational burden is reduced further by the clustering algorithm. According to the parameters of IEEE39 New England system, the geographic wiring diagram of the cyber-physical system is established. Simulation results show the proposed restoration strategy can consider the support function of the information system and space-time coordination of equipment maintenance at both sides comprehensively to speed up load recovery progress.

One of solutions to address the increasing demands for data traffic by mobile users is to implement ultra dense network (UDN). UDN consists of many femtocells that are densely deployed on macrocellular communication networks. Since femtocells radius is very short and they are tightly packed, it faces interferences problem i.e., co-tier and cross-tier interferences. This paper proposes interference management technique based on clustering method for UDN multicellular communication networks at downlink transmission. It is purposed to reduce interference effects. To evaluate the proposed clustering method, two simulation scenarios have been designed; namely baseline system and system with the proposed clustering method. The scenarios applied is to randomly distribute femtocells following a uniform distribution in three macrocells areas. Through a clustering algorithm, adjacent femtocells are grouped into one cluster and assigned different frequency channels for each femtocell in that cluster. The same cluster pattern is repeated for all femtocells. Thus, interference effect is canceled within one cluster and reduced among clusters by widening the distance among femtocells that use the same frequency channels. Through simulation experiment, the proposed clustering method is evaluated and compared to the baseline system. The simulation results show that Signal to Interference plus Noise Ratio (SINR) of femtocells increases 0.44 dB, throughput increases 1.67%, Bit Error Rate (BER) reduces, and other parameters improve as well which include spectral efficiency, network energy, and average network energy. The proposed clustering method increases performances of the networks and provides better solution for transmission data speed in densely femtocell network.

This research aims to obtain clustering results from the K-Means and K-Medoids methods for grouping communities per sub-district in Bojonegoro Regency based on several types of Social Assistance. The method used is one of the data mining methods, namely the clustering method. As a comparison, the clustering methods used are the K-Means and K-Medoids methods. The research variable used is annual data on Family Social Assistance from the Bojonegoro Regency Social Service from 2020 to 2021, which contains several types of assistance, namely Regional Contribution Assistance, Regional Direct Assistance, Village Direct Assistance, Social Food Assistance Affected by the COVID-19 Pandemic, and the Family Hope Program in the form of discrete data with a ratio scale. The results obtained by modifying the Elbow method obtained the best clustering method is the K-Means method with 5 optimal clusters. The sub-district clusters in 2020 and 2021 are different because in 2020 there is more focus on COVID-19 pandemic assistance, while in 2021, there is more focus on direct Regional and Village assistance

The development of chiral compounds with enhanced chiroptical properties is an important challenge to improve device applications. To that end, an optimization of the electric and magnetic dipole transition moments of the molecule is necessary. Nevertheless, the relationship between chemical structure and such quantum mechanical properties is not always clear. That is the case of magnetic dipole transition moment (m) for which no general trends for its optimization have been suggested. In this work we propose a general rationalization for improving the magnitude of m in different families of chiral compounds. Performing a clustering analysis of hundreds of transitions, we have been able to identify a single group in which |m| value is maximized along the helix axis. More interestingly, we have found an accurate linear relationship (up to R2=0.994) between the maximum value of this parameter and the area of the inner cavity of the helix, thus resembling classical behavior of solenoids. This research provides a tool for the rationalized synthesis of compounds with improved chiroptical responses. [ABSTRACT FROM AUTHOR]

Copyright of Machine Tool & Hydraulics is the property of Guangzhou Mechanical Engineering Research Institute (GMERI) and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Indonesia needs a daily monitoring system due to its frequent fires and, more importantly, to assist stakeholders in the field in taking action to mitigate disasters. Our method simplified the number of hotspots for field-based purposes and was verified by comparing the point-based (point-HS) VIIRS (Visible Infrared Imaging Radiometer Suite) 375m-derived temperature anomalies (hotspots) and clustered-based hotspots (cluster-HS, our suggested method). Using Euclidean clustering, we calculated the distance between hotspot points and applied specific criteria to reduce the number of hotspots while aligning them closely with fire incidents. We evaluated accuracy at different fire sizes, burned areas, peatlands, and distances from the reported burn center. We found that the accuracy increases at 1.5 km from the center of the fire for both point- and cluster-HS at 52% and 53%, respectively. For areas larger than 14 ha, both types of hotspots yielded superior results of 83%. Cluster-HS performs better on peatlands than non-peatlands (62% vs. 57%). Without diminishing the precision of the hotspot observation, this study indicates that our method is reliable for assisting field stakeholders in the field in taking actions. Therefore, this product could be implemented into Indonesia's daily hotspot monitoring. [ABSTRACT FROM AUTHOR]

Inland shipping is of great significance in economic development, and ship surveillance and classification are of great importance for ship management and dispatch. For river ship detection, ultrahigh-frequency (UHF) radar is an effective equipment owing to its wide coverage and easy deployment. The extension in range, Doppler, and azimuth and target recognition are two main problems in UHF ship detection. Clustering is a necessary step to get the center of an extended target. However, it is difficult to distinguish between different target echoes when they overlap each other in range, Doppler, and azimuth and so far practical methods for extended target recognition with UHF radar have been rarely discussed. In this study, a two-stage target classification method is proposed for UHF radar ship detection. In the first stage, grid-based gradient descent (GBGD) clustering is proposed to distinguish targets with three-dimensional (3D) information. Then in the second stage, the inverse synthetic aperture radar (ISAR) imaging algorithm is employed to differentiate ships of different types. The simulation results show that the proposed method achieves a 20% higher clustering accuracy than other methods when the targets have close 3D information. The feasibility of ISAR imaging for target classification using UHF radar is also validated via simulation. Some experimental results are also given to show the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Research purpose. Identify the main textile exporters in the global world that would have the greatest impact on the development of geared textiles.

We evaluated the genetic diversity, coancestry and heritability of an E. benthamii trial. The 115 individuals were genotyped (13 SSR) and had their height and diameter at breast height (dbh) measured. Heritability was estimated using the RR-BLUP and the pairwise kinship coefficient method. An average of nine alleles per locus was observed. The expected heterozygosity (0.655) was similar to the observed heterozygosity, with the estimated inbreeding (0.02) being low. The group coancestry (0.051) demonstrates that the trees are related to some degree. The trees were clustered in five groups using the Evanno's method. The average kinship within each group ranges from 0.042 to 0.082. The heritability estimated by RR-BLUP was low. The heritability estimated using the kinship coefficients is moderate, reaching estimated genetic gains of 14% for dbh. After knowing how genetic groups are distributed within the population, strategies for collecting, conserving, and using these germplasm resources can be performed. [ABSTRACT FROM AUTHOR]

Clustering, an unsupervised machine learning technique, plays a crucial role in partitioning unlabeled data into meaningful groups. K-means, known for its simplicity, has gained popularity as a clustering method. However, both K-means and the LAC algorithm, which utilize learning automata, are sensitive to the selection of initial points. To overcome this limitation, we propose an enhanced LAC algorithm based on the K-Harmonic means approach. We evaluate its performance on seven datasets and demonstrate its superiority over other representative algorithms. Moreover, we tailor this algorithm to address the controller placement problem in software-defined networks, a critical field in this context. To optimize relevant parameters such as switch–controller delay, intercontroller delay, and load balancing, we leverage learning automata. In our comparative analysis conducted in Python, we benchmark our algorithm against spectral, K-means, and LAC algorithms on four different network topologies. The results unequivocally show that our proposed algorithm outperforms the others, achieving a significant improvement ranging from 3 to 11 percent. This research contributes to the advancement of clustering techniques and their practical application in software-defined networks. [ABSTRACT FROM AUTHOR]

In wireless sensor networks (WSNs), balancing energy consumption is always a critical problem. Multi-hop data transmission may cause the nodes around the static sink to exhaust energy prematurely. It can be avoided by introducing the mobile sink (MS) in WSNs. The MS only needs to visit some specific nodes, such as cluster heads (CHs) in the clustering network, to collect data. In this paper, we propose an energy-adaptive clustering method based on Taguchi-based-GWO optimizer (EACM-TGWO). Different from the existing clustering protocols, we develop the optimal number of CHs according to the characteristic of energy consumption in MS-based WSNs. Afterwards, the fitness function is established by combining the residual energy and the average distance within clusters to select CHs. Meanwhile, an adaptive weight factor is introduced to dynamically tune the influence degree of the two factors on the clustering results. Furthermore, a novel meta-heuristic algorithm Taguchi-based grey wolf optimizer (TGWO) is used to search for the optimal CHs set. Compared with grey wolf optimizer (GWO) and some of its improved versions, TGWO is more excellent tested in CEC2017. We also make an elaborate comparison of EACM-TGWO with LEACH-C, EACM-GWO, MSECA, and GAOC. The simulation results indicate that EACM-TGWO is more powerful in terms of balancing energy consumption and saving network energy. [ABSTRACT FROM AUTHOR]

In cylinder/wall interaction systems, the evolution of secondary vortex structures causes the early transition of the boundary layer. Investigations of secondary vortex statistics can enhance our understanding of the complex interactions between blunt body wakes and wall-bounded shear layers. In this paper, the statistical properties of three-dimensional vortex structures in the near-wake region during cylinder/wall interactions are numerically simulated by large-eddy simulations (LES) at the gap ratio G/D=1.0. We use a clustering method to identify complex vortex structures associated with intense events. According to the volume distribution of the minimum and maximum wall distances of these vortex structures, ymin and ymax, the spatial population of vortex clusters can be divided into two groups: wake vortex region and secondary vortex region. Secondary vortex structures are found in the second region. Self-similar secondary vortex structures have power law relationships. The deflection patterns of secondary and wake vortex structures are visualized by the joint PDFs of the vortex centre positions. The upwards deflection of the secondary vortex is greater than that of the lower wake vortex. The lead-lag phenomenon of secondary and lower wake vortices is observed through the temporal dynamics of the domain integral enstrophy. The conditional average results demonstrate that high-speed structures u ′ ˜ > 0 appear above the secondary vortex due to the influence of the lower wake vortex. [ABSTRACT FROM AUTHOR]

In recent years, solid-state drives (SSDs) that adopt NAND flash memory have been widely used as the main storage devices. In particular, NAND flash memory has a special feature of “out-of-place” updates to write the up-to-date data to a free page, and the corresponding old page will become invalid. When the number of free pages in SSDs is insufficient, garbage collection (GC) will be executed to reclaim the invalid pages in a block by erasing the block. Many studies have shown that a good hot/cold data separation (i.e., clustering) can greatly reduce the overhead of GC so as to improve the SSD performance. However, previous clustering methods usually use a static number of clusters or a fixed size of granularity (i.e., a fine-grained or a coarsegrained granularity), so they may not always perform well for different kinds of workloads. Therefore, we propose a granularity-based clustering method to adaptively adjust the size of granularity groups for an appropriate number of clusters at runtime according to the update distances of logical addresses. According to the experimental results, we can improve SSD performance by reducing the overhead of GC and decrease the write amplification. Furthermore, we can show that the proposed method can utilize a fine-grained granularity to retain the record accuracy of update distances and also utilize a coarse-grained granularity to reduce the space required to record update distances. [ABSTRACT FROM AUTHOR]

The unsupervised domain adaptive (UDA) person re-identification (re-ID) method is of great significance to promote the practical application of person re-ID. However, the noisy pseudo labels in the target domain hinder its performance. In this paper, a novel high-quality pseudo labels (HQP) method for UDA person re-ID is proposed, which improves the performance from the perspectives of sample feature expression and similarity measurement in the clustering. In order to obtain better feature representation for target domain samples, a source domain generalization method based on contrastive learning (SCL) is designed. SCL learns the inherently consistent information within a sample, thereby improving the expression ability of the source domain pre-trained model. In order to provide a more reasonable similarity measurement for the clustering method, a soft label similarity based on neighborhood information integration (NII) is designed, which aids the clustering method to generate reliable pseudo labels. Market-1501, DukeMTMC-ReID and MSMT17 datasets are employed to evaluate the performance of the proposed HQP method. It achieves the results of 80.3%/92.3%, 68.0%/82.6% and 25.4/53.3 mAP/Rank-1 on DukeMTMC-ReID-to-Market-1501, Market-1501-to-DukeMTMC-ReID and DukeMTMC-ReID-to-MSMT17 tasks. Experimental results demonstrate that our HQP method performs favorably against the state-of-the-art UDA person re-ID methods. [ABSTRACT FROM AUTHOR]

Environment perception is a key part of robot self-controlled motion. When using vision to accomplish obstacle detection tasks, it is difficult for deep learning methods to detect all obstacles due to complex environment and vision limitations, and it is difficult for traditional methods to meet real-time requirements when applied to embedded platforms. In this paper, a fast obstacle-detection process applied to RGB-D cameras is proposed. The process has three main steps, feature point extraction, noise removal, and obstacle clustering. Using Canny and Shi–Tomasi algorithms to complete the pre-processing and feature point extraction, filtering noise based on geometry, grouping obstacles with different depths based on the basic principle that the feature points on the same object contour must be continuous or within the same depth in the view of RGB-D camera, and then doing further segmentation from the horizontal direction to complete the obstacle clustering work. The method omits the iterative computation process required by traditional methods and greatly reduces the memory and time overhead. After experimental verification, the proposed method has a comprehensive recognition accuracy of 82.41%, which is 4.13% and 19.34% higher than that of RSC and traditional methods, respectively, and recognition accuracy of 91.72% under normal illumination, with a recognition speed of more than 20 FPS on the embedded platform; at the same time, all detections can be achieved within 1 m under normal illumination, and the detection error is no more than 2 cm within 3 m. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach for measuring the human respiration rate utilizing a mattress sensor. The proposed method employs an unsupervised clustering technique in conjunction with a cluster selection algorithm to accurately measure the respiration rate. The pressure exerted on the mattress sensor reflects the volume changes in the upper body during the breathing process, enabling the estimation of the respiration waveform based on the measured pressure variations. Through the clustering method, the measured pressure changes are effectively separated into respiratory-related clusters and noise. Subsequently, the cluster selection algorithm identifies the optimal combination of clusters that best represents the respiration pattern. To evaluate the performance of the proposed method, it was compared against two alternative methods, namely RWD and RAC. The results demonstrate that the proposed method, referred to as RCS, achieved the highest mean Pearson correlation coefficient of 0.88, indicating superior performance compared to the other methods. In contrast, the mean Pearson correlation coefficients for RWD and RAC were determined as 0.61 and 0.76, respectively. Statistical analysis confirmed the significance of the differences among these methods (p < 0.001). Furthermore, a regression analysis was conducted to assess the accuracy of the respiration rate measurements. The findings revealed that the proposed method yielded the most precise estimation of the respiratory rate.

Federated learning is a distributed machine learning framework that enables a large number of devices to cooperatively train a model without data sharing. However, because federated learning trains a model using non-independent and identically distributed (non-IID) data stored at local devices, the weight divergence causes a performance loss. This paper focuses on solving the non-IID problems and proposes Kalman filter-based clustering federated learning method called K-FL to get performance gain by providing a specific model with low variance to the device. To the best of our knowledge, it is the first clustering federated learning method that can train a model requiring fewer communication rounds under the premise that non-IID environment without any prior knowledge and an initial value set by the user. From simulations, we demonstrate that the proposed K-FL can train a model much faster, requiring fewer communication rounds than FedAvg and LG-FedAvg when testing neural networks using the MNIST, FMNIST, and CIFAR-10 datasets. As a numerical result, it is shown that the accuracy is improved in all datasets while the computational time cost is reduced by $1.43\times $ , $1.67\times $ , and $1.63\times $ compared to FedAvg, respectively.

ABSTRACT The cultivation of Cotton (Gossypium hirsutum) is trending in an upward expansion at Minas Gerais state, Brazil. Due this growth, the breeding program has been demanded for the knowledge about the genetic diversity and correlation between morpho-agronomic traits, to start interesting crosses. The objective of this work was to study the genetic diversity and the correlation network between morpho-agronomic traits of cotton genotypes. Two cotton accessions and five commercial cultivars: FM993, FMT701, FM910, DP604 and DP604BG were evaluated for 16 traits. The results revealed divergence between the genotypes. Unweighted pair-group method arithmetic average (UPGMA) analysis allocated the genotypes in three clusters, with DP604 and Accession 2 being the most divergent. The correlation network helped to visualize the association between traits, presenting a negative correlation between anthracnose with plant height (-0.465), plant vigor (-0.486) and main root length (-0.437). Also, even with some genotypes presenting a higher number of reproductive branches, the correlation for this trait with number of bolls showed a high significant value (0.68). Thus, the genotypes showed wide genetic diversity for the composition of future crosses in cotton breeding programs. The correlation network allowed the visualization of a medium to high correlation pattern for the morpho-agronomic traits.

Collaborative filtering recommends items on the principle of similarity. The effectiveness of the similarity is highly dependent on the performance of the clustering algorithm. This paper presents a new clustering-based collaborative filtering method for efficient item recommendation. The new approach introduces a new variant of the grasshopper optimization algorithm to determine optimal similarity patterns. Nine nature-inspired algorithms have been considered and tested against seventeen standard benchmark functions in terms of mean fitness value and the Friedman test to validate the proposed variant experimentally. Further, the efficacy of the new clustering-based collaborative filtering method is tested on the publicly available MovieLens dataset and evaluated in terms of mean absolute error, precision, and recall over the different cluster settings. The obtained results are compared against nine clustering-based collaborative filtering methods. Experiments affirm that the proposed method is a permissive technique for the recommendation on large-scale datasets. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Tiangong University is the property of Journal of Tianjin Polytechnic University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

With the development and maturity of 5G communication technology, the intelligent terminals and applications in the Internet of Things have exploded. The large-capacity data generated by the Internet of Things has higher transmission requirements on the network. Opportunistic social networks use the "store-carry-forward" method to complete message delivery, so it is particularly important to select reliable relay users to ensure the integrity of data transmission. However, users have limited memory resources. A large number of redundant message copies in the network will lead to increased node energy consumption, which will further cause network congestion. In addition, the user's movement is random and dynamic. Frequent changes in the network topology may lead to transmission link interruption and data loss. Based on the characteristics of social networks, this paper proposes a clustered dataset transmission strategy in opportunistic social networks based on the probability of encounters between users and the characteristics of user motion (PC-OSN). First, the method divides users into different clusters according to the strength of the relationship between users, which effectively reduces the occupation of network resources when users are clustered. Then PC-OSN adopts different message forwarding judgment indicators within and between clusters. PC-OSN forwards messages in a cluster by using binary injection to ensure that messages are delivered to users with a high probability of encountering. The user's activity level and motion characteristics are considered comprehensively among the clusters to determine the user's priority so as to deliver the message to the reliable user. PC-OSN fully considers the characteristics of users and the correlation between users, reducing the blindness of message forwarding and waste of ineffective resources. Experimental results show that PC-OSN achieves a message delivery rate of 0.94, which is about 14% higher than other methods. Its average latency and network overhead are reduced by about 24% and 16%, respectively. [ABSTRACT FROM AUTHOR]

The construction industry has the highest frequency of injuries and fatal accidents compared to other industries. Therefore, an accurate evaluation of the level of risk is necessary to effectively manage accidents in the construction industry. Although risk assessments are used in the construction industry to identify and reduce risk, they are carried out qualitatively by the subjective opinion of safety managers. Therefore, this study proposed a quantitative risk assessment based on the risk levels of 17 building facilities. Moreover, this study classified the building facilities into groups of similar risk levels. This study involved five steps. The results of the study indicated that in terms of both accident probability and financial losses, the facility type with the highest risk ranking was "Residential buildings" (5.92). The risk assessment method proposed in this study can contribute toward reducing accidents by identifying the exact risk level for each facility type. Moreover, clients and contractors can utilize this study and its findings as a basis for different safety management approaches depending on the risk group level. [ABSTRACT FROM AUTHOR]

Indonesia needs a daily monitoring system due to its frequent fires and, more importantly, to assist stakeholders in the field in taking action to mitigate disasters. Our method simplified the number of hotspots for field-based purposes and was verified by comparing the point-based (point-HS) VIIRS (Visible Infrared Imaging Radiometer Suite) 375m-derived temperature anomalies (hotspots) and clustered-based hotspots (cluster-HS, our suggested method). Using Euclidean clustering, we calculated the distance between hotspot points and applied specific criteria to reduce the number of hotspots while aligning them closely with fire incidents. We evaluated accuracy at different fire sizes, burned areas, peatlands, and distances from the reported burn center. We found that the accuracy increases at 1.5 km from the center of the fire for both point- and cluster-HS at 52% and 53%, respectively. For areas larger than 14 ha, both types of hotspots yielded superior results of 83%. Cluster-HS performs better on peatlands than non-peatlands (62% vs. 57%). Without diminishing the precision of the hotspot observation, this study indicates that our method is reliable for assisting field stakeholders in the field in taking actions. Therefore, this product could be implemented into Indonesia’s daily hotspot monitoring.

Inland shipping is of great significance in economic development, and ship surveillance and classification are of great importance for ship management and dispatch. For river ship detection, ultrahigh-frequency (UHF) radar is an effective equipment owing to its wide coverage and easy deployment. The extension in range, Doppler, and azimuth and target recognition are two main problems in UHF ship detection. Clustering is a necessary step to get the center of an extended target. However, it is difficult to distinguish between different target echoes when they overlap each other in range, Doppler, and azimuth and so far practical methods for extended target recognition with UHF radar have been rarely discussed. In this study, a two-stage target classification method is proposed for UHF radar ship detection. In the first stage, grid-based gradient descent (GBGD) clustering is proposed to distinguish targets with three-dimensional (3D) information. Then in the second stage, the inverse synthetic aperture radar (ISAR) imaging algorithm is employed to differentiate ships of different types. The simulation results show that the proposed method achieves a 20% higher clustering accuracy than other methods when the targets have close 3D information. The feasibility of ISAR imaging for target classification using UHF radar is also validated via simulation. Some experimental results are also given to show the effectiveness of the proposed method.

The aim of this study is built in two phases: to quantify the ability of novel milk metabolites to measure between-animal variability in response and recovery profiles to a short-term nutritional challenge, then to derive a resilience index from the relationship between these individual variations. At two different stages of lactation, sixteen lactating dairy goats were exposed to a 2-d underfeeding challenge. The first challenge was in late lactation, and the second was carried out on the same goats early in the following lactation. During the entire experiment period, samples were taken at each milking for milk metabolite measures. For each metabolite, the response profile of each goat was characterised using a piecewise model for describing the dynamic pattern of response and recovery profiles after the challenge relative to the start of the nutritional challenge. Cluster Analysis identified three types of response/recovery profiles per metabolite. Using cluster membership, multiple correspondence analyses (MCAs) were performed to further characterise response profile types across animals and metabolites. This MCA analysis identified three groups of animals. Further, discriminant path analysis was able to separate these groups of multivariate response/recovery profile type based on threshold levels of three milk metabolites: β-hydroxybutyrate, free glucose and uric acid. Further analyses were done to explore the possibility of developing an index of resilience from milk metabolite measures. Different types of performance response to short-term nutritional challenge can be distinguished using multivariate analyses of a panel of milk metabolites.

An important topic in reservoir studies is to separate different reservoir compartments. Recently, the Fourier Transform Infrared (FTIR) analysis has been regarded for this purpose. Developed in the Persian Gulf, southern Iran, the Yammama Formation is a thick carbonate reservoir that has been dated back to the Lower Cretaceous. Dealing with the compartmentalization problem of Yammama has been a challenging task due to its heterogeneous nature as a major fault cuts through the reservoir. In this research, eleven oil samples were collected from various wells penetrating the Yammama Reservoir to classify its compartments using FTIR and multivariate statistical methods. Five different indexes were extracted from the FTIR analysis diagrams. These represented the aliphatic content, aromatic content, long-chain compounds, and substitution 1 and 2. The extracted information was used as input for hierarchical and k-means clustering methods. The obtained results showed that the Yammama Reservoir comprises two distinguished compartments in terms of the oil chemical composition. The wells located in the northwestern and southeastern parts of the study area were classified under two different clusters. The obtained compartmentalization from the FTIR analysis and the clustering methods was in good agreement with previous findings. [ABSTRACT FROM AUTHOR]

Unweighted Pair Group Method with Arithmetic Mean (UPGMA) is one of the most popular distance-based methods to reconstruct an equidistant phylogenetic tree from a distance matrix computed from an alignment of sequences. Since we use equidistant trees as gene trees for phylogenomic analyses under the multi-species coalescent model and since an input distance matrix computed from an alignment of each gene in a genome is estimated via the maximum likelihood estimators, it is important to conduct a robust analysis on UPGMA. Stochastic safety radius, introduced by Steel and Gascuel, provides a lower bound for the probability that a phylogenetic tree reconstruction method returns the true tree topology from a given distance matrix. In this article, we compute the stochastic safety radius of UPGMA for a phylogenetic tree with n leaves. Computational experiments show an improved gap between empirical probabilities estimated from random samples and the true tree topology from UPGMA, increasing confidence in phylogenic results. [ABSTRACT FROM AUTHOR]

Soil depth, a major parameter influencing the behavior of landslides, was estimated by scanning rocky surfaces in mountainous areas to spatially obtain the thickness of the soil zone. The emphasis was on quantifying the rocky surface identification with the soil depth estimation by using the K-means clustering method and geographic information system (GIS) data on topology, land use, rock proportion, and satellite images. The results of the rocky surface identification in the study area were compared with the field survey, and it was shown that the classification with and without rocky surfaces in the mountainous area was different from the other natural mountains with basal soil areas only. Based on the results obtained, a simple method is presented for identifying the spatial thickness of the soil zone in mountainous areas by comparison with in situ borehole data. Moreover, a landslide-debris flow simulation was performed using the estimated soil depth map, and it was found that the prediction results using the proposed method showed relatively high accuracy compared with other predictions overestimated by excessive soil depth. [ABSTRACT FROM AUTHOR]

The cultivation of Cotton (Gossypium hirsutum) is trending in an upward expansion at Minas Gerais state, Brazil. Due this growth, the breeding program has been demanded for the knowledge about the genetic diversity and correlation between morpho-agronomic traits, to start interesting crosses. The objective of this work was to study the genetic diversity and the correlation network between morpho-agronomic traits of cotton genotypes. Two cotton accessions and five commercial cultivars: FM993, FMT701, FM910, DP604 and DP604BG were evaluated for 16 traits. The results revealed divergence between the genotypes. Unweighted pair-group method arithmetic average (UPGMA) analysis allocated the genotypes in three clusters, with DP604 and Accession 2 being the most divergent. The correlation network helped to visualize the association between traits, presenting a negative correlation between anthracnose with plant height (-0.465), plant vigor (-0.486) and main root length (-0.437). Also, even with some genotypes presenting a higher number of reproductive branches, the correlation for this trait with number of bolls showed a high significant value (0.68). Thus, the genotypes showed wide genetic diversity for the composition of future crosses in cotton breeding programs. The correlation network allowed the visualization of a medium to high correlation pattern for the morpho-agronomic traits. [ABSTRACT FROM AUTHOR]

Abstract The financial industry is a key to promoting the development of the national economy, and the risk it takes is also the largest hidden risk in the financial market. Therefore, the risk existing in the current financial market should be deeply explored under blockchain technology (BT) to ensure the functions of financial markets. The risk of financial markets is analyzed using machine learning (ML) and random forest (RF). First, the clustering method is introduced, and an example is given to illustrate the RF classification model. The collected data sets are divided into test sets and training sets, the corresponding rules are formulated and generated, and the branches of the decision tree (DT) are constructed according to the optimization principle. Finally, the steps of constructing the branches of DT are repeated until they are not continued. The results show that the three major industries of the regional economy account for 3.5%, 51.8%, 3.2%, 3.4%, and 3.8% of the regional GDP, respectively, the secondary industry makes up 44.5%, 43%, 45.1%, 44.8%, and 43.6%, respectively, and the tertiary industry occupies 20%, 3.7%, 52.3%, 52.9%, 54%, and 54.6%, respectively. This shows that with the development of the industrial structure under BT, the economic subject gradually shifts from the primary industry to the tertiary industry; BT can improve the efficiency of the financial industry and reduce operating costs and dependence on media. Meanwhile, the financial features of BT can provide a good platform for business expansion. The application of BT to the supply chain gives a theoretical reference for promoting the synergy between companies.

There has been limited research on integrating sustainability factors into indicators of global competitiveness among nations. Extant measures of macroeconomic competitiveness focus inadequate attention on the relative importance and interdependences among economic, social, and environmental factors and often their aggregation method. The primary objectives of this study were to explore the interrelations of economic competitiveness and sustainability while integrating the economic and sustainability performance indicators of the Global Competitiveness Index (GCI) and the Global Sustainable Competitiveness Index (GSCI) to create more accurate rankings of countries in both dimensions. Deep Ranking Analysis by Power Eigenvectors (DRAPE) with Variable Clustering (VARCLUS) was applied to associate multiple and conflicting economic and sustainable competitiveness criteria. Consequently, economic competitiveness and mainly institutional environment and innovation capabilities tend to couple with a higher degree of sustainability. However, improving and refining sustainable competitiveness may be vulnerable to environmental issues and resource scarcity, but further socio-economic dimensions must also be considered. The outcomes demonstrate the leadership of the Scandinavian countries and Switzerland, which are all competitive as well as sustainable. This research may extend and inspire further studies and government interventions that adopt more comprehensive approaches to assessing global competitiveness.

• 12 rice phenology models were calibrated and evaluated with data from 46 sites. • Thermal requirements (TR) varied across observation sites and years for all models. • Cluster-specific TRs can achieve higher accuracy due to less heterogeneity. • Season mean temperature and latitude were the most effective clustering variables. • TR variations should not be ignored in large-scale crop modeling studies. Process-based phenological models use thermal requirement (TR) defined by planting date, temperature and photoperiod to predict crop developmental stages. The TR value for a specific developmental stage for a given variety is often presumed to be constant regardless of environmental conditions. We calibrated and compared 12 phenological models using 27-year of observation data of one unique rice (Oryza sativa L.) variety ('Shanyou63') from 46 sites in southern China. Our findings indicated that shifts in environmental conditions significantly affected TR values, e.g., standard deviations of TR at physiological maturation ranged from 83 to 167 °C d. Clustering sites together minimized environmental heterogeneity, and thus minimized the differences in TR for different phenological stages. When the increased from 1 to 24, simulation errors for the 12 models showed a significant decrease across all developmental stages, from 1.8 to 1.4 days for tillering, from 5.3 to 3.7 days for jointing, from 5.6 to 3.9 days for booting, from 4.7 to 3.3 days for heading, and from 6.6 to 3.7 days for physiological maturation. Furthermore, our findings indicate that models featuring a three-segment piecewise linear temperature response function provide a more precise prediction. In contrast, models incorporating a Beta temperature response function have not performed well. This difference is attributed to different mechanisms used to describe the response of crop development rate to temperature, particularly at non-optimum temperatures. The impact of the photoperiod response function on prediction accuracy became significant with the expansion of scale. Our results demonstrate that clustering method effectively compensates for the lack of crop adaptation processes in common phenological models, leading to significantly improved phenological prediction accuracy in regions with environmental diversity. [ABSTRACT FROM AUTHOR]

The three-dimensional structures based on the quadrant classification of the tangential Reynolds stress (Qs), are studied in direct numerical simulations of cylinder-wake/boundary-layer interaction with a moderate gap ratio (G / D = 1.0) and low or middle Reynolds numbers. In this paper, direct numerical simulations are performed using the immersed boundary lattice Boltzmann method (LBM). We utilize a three-dimensional clustering method to identify Reynolds shear stress structures (Qs -structures) associated with intense events. The kinematic characteristics of coherent structures are discussed through the statistical properties of Qs -structures. The spatial relationship of between vortex clusters and Qs -structures (mainly Q 2-structures as well as Q 4-structures) is explained by their evolution characteristics. The spatial distribution of Qs -structures, determined by the volume distribution of the minimum and maximum wall distances (y m i n and y m a x), are found to be similar to that of vortex clusters in the cylinder/wall interaction. In the near-wall statistical region, the secondary vortex clusters are surrounded by the parallel Q 2 and Q 4 pairs, and mostly lodged within the Q 2. During the evolution process of coherent structures, the Q 2-structures and the secondary vortex clusters developed into hairpin-like structures, respectively. The conditional average results demonstrated that the generated hairpin vortex structures are attributed to the momentum transfers. • The evolution of vortex clusters and Reynolds shear stress structures is good coordination. • The secondary vortex clusters and Q2-structures develop into hairpin-like structures. • The spatial distribution of Qs is similar to the vortex clusters in cylinder/wall interaction. • Q2 and Q4 tend to form parallel pairs. • It can be observed that the secondary vortex clusters are more inclined towards Q2. [ABSTRACT FROM AUTHOR]

Clustering, an unsupervised machine learning technique, plays a crucial role in partitioning unlabeled data into meaningful groups. K-means, known for its simplicity, has gained popularity as a clustering method. However, both K-means and the LAC algorithm, which utilize learning automata, are sensitive to the selection of initial points. To overcome this limitation, we propose an enhanced LAC algorithm based on the K-Harmonic means approach. We evaluate its performance on seven datasets and demonstrate its superiority over other representative algorithms. Moreover, we tailor this algorithm to address the controller placement problem in software-defined networks, a critical field in this context. To optimize relevant parameters such as switch–controller delay, intercontroller delay, and load balancing, we leverage learning automata. In our comparative analysis conducted in Python, we benchmark our algorithm against spectral, K-means, and LAC algorithms on four different network topologies. The results unequivocally show that our proposed algorithm outperforms the others, achieving a significant improvement ranging from 3 to 11 percent. This research contributes to the advancement of clustering techniques and their practical application in software-defined networks.

In this era, research and development are becoming a continuous and accelerating process because technology changes rapidly with a short lifecycle. As a result, various methodologies are being developed to monitor these rapidly changing research trends; In particular, clustering method-related studies in science and technology documents are being developed with a variety of approaches. However, previous studies on document clustering methods focus on a specific field or language but do not take into consideration certain important pieces of information in science and technology documents. Therefore, this study proposes an embedding methodology that uses important content from scientific and technical documents. We took into consideration the importance of information containing core structures in science and technology documents and proposed a clustering methodology that analyzes structured and unstructured data, such as textual information, author information, and citation information. The proposed method combines both textual and structural data from the paper, using a method that focuses on screening important information by sections in science and technology documents. Then, Girvan-Newman clustering and Louvain clustering models are applied to generate embedding vectors and show evaluation results through the clustering indices. As a practical example, we applied the proposed methodology using paper data from the field of hydrogen cell vehicles. The results of this study will be effective in identifying gaps in technology for new technological development, identifying technology trends, and presenting directional information for future technology development.