Dynamic Clustering Research Articles

Meta-learning on dynamic node clustering knowledge graph for cold-start recommendation

Meta-learning has been introduced in the recommendation domain, and a possible direction to extend graph-based meta-learning is how to exploit higher-order information between nodes. The current studies primarily rely on pre-embedding methods to utilize user attributes, inadvertently overlooking the crucial co-occurrence patterns among users’ features. This oversight hinders the discovery of implicit connections between users. Furthermore, in scenarios of infrequent interactions, the reliance on initial edge information constrains connections between nodes in the same category. This constraint compromises the update capability of node embeddings, leading to reduced recommendation accuracy. In this paper, a meta-learning recommendation method with a dynamic node clustering knowledge graph(DCKG) is proposed. By introducing user attribute nodes and a triple-gated attention networks, this approach facilitates connections between users based on attribute relationships and enables the balancing of information across various pathways. Additionally, the integration of a dynamic aggregation module frees nodes of the same category from initial connection constraints, fostering more direct connections and further augmenting the propagation capability of higher-order information. The experimental results show that DCKG achieves the excellent results in the cold-start recommendation. DCKG can extend the use of higher-order features in the graph by direct connection.

Cluster dynamics study on nano damage of RPV steels under proton irradiation at 290°C

Irradiation-induced defects such as dislocation loops, cavities or solute clusters and chemical composition segregation of reactor pressure vessel (RPV) steel are the root causes of irradiation embrittlement. Combining two nucleation mechanisms, namely, the uniform nucleation and non-uniform nucleation of solute clusters (such as Cu-rich phase), a cluster kinetic simulation was established based on the reaction rate theory, and the co-evolution of matrix damage and Cu-rich phase in low-copper RPV steel was simulated under irradiation. And the average size and number density of defective clusters and solute clusters were established with irradiation dose. Compared with the average size and number density of dislocation loops observed by transmission electron microscopy (TEM) of proton irradiated RPV steel at 290°C, the verification results show that the cluster dynamics model considering both the nucleation mechanism of interstitial dislocation loops and vacancy clusters can well simulate the irradiation damage behavior of materials.

Dynamic clustering of Asian regionalisms: Empirical testing of United Nations multilateral treaty participation by social network analysis

This paper investigates the evolving clustering and historical progression of “Asian regionalisms” concerning their involvement in multilateral treaties deposited in the United Nations system. We employ criteria such as geographic proximity, historical connections, cultural affinities, and economic interdependencies to identify twenty-eight candidate countries from East Asia, Southeast Asia, South Asia, and Central Asia for this empirical testing. Using a social network analysis approach, we model the network of these twenty-eight Asian state actors alongside 600 major treaties from the United Nations system, identifying clusters among Asian states by assessing similarities in their treaty participation behavior. Specifically, we observe dynamic changes in these clusters across three key historical eras: Post-war reconstruction and transformation (1945–1968), Cold War tensions and global transformations (1969–1989), and post-Cold War era and globalization (1990–present). Employing the Louvain cluster detection algorithm, the results reveal the evolution in cluster numbers and changes in membership status throughout the world timeline. The results also identify the current situation of six distinct Asian clusters based on states’ inclinations to engage or abstain from multilateral treaties across six policy domains. These findings provide a foundation for further research on the trajectories of Asian regionalisms amidst evolving global dynamics and offer insights into potential alliances, cooperation, or conflicts within the region.

Are heterogeinity and conflicting preferences no longer a problem? Personality-based dynamic clustering for group recommender systems

The complexity associated with groups of tourists led to the emergence of Group Recommender Systems (GRS) for tourism. But if generating recommendations for small groups is a complex task, to provide them to large and occasional groups is even more. This complexity is especially due to the group’s heterogeinity, conflicting preferences, the information overload found on the internet and the tourists’ different ways of coping with the information, hindering the recommendation process from the users’ profile construction to the final recommendation of a list of points of interest to visit. In this work, we show how we tackled the identified issues in a GRS prototype, Grouplanner, including the cold-start problem, by predicting the tourists’ preferences based only on their personality and dividing the main group into subgroups of similar personality; by using a Multi-Agent Microservice; a novel dynamic clustering algorithm, d-means, adapted from the k-means algorithm, that does not need to know the number of clusters a priori; and association rules. Using a personality dataset of n=100k users, the proposed d-means algorithm was tested against two baselines (k-means and k-means++), showing better results in the clustering quality and scalability. We were also able to determine a large set of association rules to refine the recommendations, although further improvements are needed. To test the Grouplanner prototype, a simulation with real users (n=35) was conducted. The results showed the subgroups formed were very compact, revealing a very good clustering quality, with an average silhouette of s = 0.91. 11 of the 15 proposed tourist preferences were successfully predicted and used for the preliminary recommendation lists, being 92 % of the participants satisfied with the individual recommendations and 96 % with the group recommendations.

On the Arrival Time Distribution of Reacting Systems in Ion Mobility Spectrometry.

Ion mobility spectrometry (IMS) is a widely used gas-phase separation technique, particularly when coupled with mass spectrometry (MS). Modern IMS instruments often apply elevated reduced field strengths for improved ion separation and ion focusing. These alter the collision dynamics and further drive ion reaction processes that can change the analyte's structure. As a result, the measured arrival time distribution (ATD) can change with the applied reduced field strengths. In this work, we systematically study how the ion collision dynamics and the ion reaction dynamics, as a function of the reduced field strength, can alter the ATD. To this end, we investigate 2,6-di-tert-butylpyridine, methanol, and ethyl acetate using a home-built drift tube IMS coupled to a home-built MS and extensive first-principles Monte Carlo modeling. We show how elevated reduced field strengths can actually lower resolving power through increased ion diffusion and how the field dependency of the ion mobility can introduce uncertainties to collision cross sections (CCS) calculated from the measured mobilities. On top of the collision dynamics, we show how chemical transformation processes that alter the analyte's CCS, e.g., dynamic clustering or fragmentation, can lead to broadened, shifted, or non-Gaussian ATDs and how sensitive these processes are to the applied field strengths. We highlight how first-principles ion dynamics simulations can help to understand and even harness the mentioned effects.

Technique for Kernel Matching Pursuit Based on Intuitionistic Fuzzy c-Means Clustering

Kernel matching pursuit (KMP) requires every step of the searching process to be global optimal searching in the redundant dictionary of functions in order to select the best matching signal structure. Namely, the dictionary learning time of KMP is too long. To solve the above drawbacks, a rough dataset was divided into some small-sized dictionaries to substitute local searching for global searching by using the property superiority of dynamic clustering performance, which is also superior in the intuitionistic fuzzy c-means (IFCM) algorithm. Then, we proposed a novel technique for KMP based on IFCM (IFCM-KMP). Subsequently, three tests including classification, effectiveness, and time complexity were carried out on four practical sample datasets, the conclusions of which fully demonstrate that the IFCM-KMP algorithm is superior to FCM and KMP.

Dcaro: Dynamic cluster formation and AUV-aided routing optimization for energy-efficient UASNs

Dcaro: Dynamic cluster formation and AUV-aided routing optimization for energy-efficient UASNs

A Review of Cluster Dynamics in Studying Radiation Damage: Dominant Factors and Practical Implications

A Review of Cluster Dynamics in Studying Radiation Damage: Dominant Factors and Practical Implications

Topology reduction through machine learning to accelerate dynamic simulation of district heating

District heating networks (DHNs) provide an efficient heat distribution solution in urban areas, accomplished through interconnected and insulated pipes linking local heat sources to local consumers. This efficiency is further enhanced by the capacity of these networks to integrate renewable heat sources and thermal storage systems. However, integration of these systems adds complexity to the physical dynamics of the network, necessitating complex dynamic simulation models. These dynamic physical simulations are computationally expensive, limiting their adoption, particularly in large-scale networks. To address this challenge, we propose a methodology utilizing Artificial Neural Networks (ANNs) to reduce the computational time associated with the DHNs dynamic simulations. Our approach consists in replacing predefined clusters of substations within the DHNs with trained surrogate ANNs models, effectively transforming these clusters into single nodes. This creates a hybrid simulation framework combining the predictions of the ANNs models with the accurate physical simulations of remaining substation nodes and pipes. We evaluate different architectures of Artificial Neural Network on diverse clusters from four synthetic DHNs with realistic heating demands. Results demonstrate that ANNs effectively learn cluster dynamics irrespective of topology or heating demand levels. Through our experiments, we achieved a 27% reduction in simulation time by replacing 39% of consumer nodes while maintaining acceptable accuracy in preserving the generated heat powers by sources.

Analyzing foreland dynamics in China's port clusters under global major events (2019–2022) by AIS trajectory data

Port foreland, serves as the maritime mirror of the hinterland, encompasses the ports and overseas markets interconnected through shipping services. During major global events, understanding port foreland structures becomes imperative for nations and port authorities to anticipate future trade trends and potential coastal management risks. However, systematic approaches for accurately identifying foreland structures remain lacking, particularly given liner shipping complexities. In response, this paper develops two algorithms to analyze changes in the foreland structure of five port clusters in China for the years 2019 and 2022, leveraging over 2.1 billion global vessel AIS trajectory signals. We aim to understand the foreland dynamics of China's ports amidst the broader impacts of the China-US trade war, the COVID-19 pandemic, and the Russia-Ukraine (RU) conflict, and provide insights for maintaining the global trade continuity. Our findings show that geopolitical events and bilateral relations highly influence port foreland structures, yet global supply chain requirements prompt stakeholders to adopt alternative transport strategies. In 2022, China's foreland structures remain stable despite its strict epidemic control measures. Positive developments in China's port forelands with the US are driven by improving bilateral relations and reduced impacts of the US-China trade war, while conflicts in RU cause shrinkage in Europe. On this basis, we provide three constructive opinions on the development of China's port clusters.

PP2A-B55 phosphatase counteracts Ki-67-dependent chromosome individualization during mitosis

Cell cycle progression is regulated by the orderly balance between kinase and phosphatase activities. PP2A phosphatase holoenzymes containing the B55 family of regulatory B subunits function as major CDK1-counteracting phosphatases during mitotic exit in mammals. However, the identification of the specific mitotic roles of these PP2A-B55 complexes has been hindered by the existence of multiple B55 isoforms. Here, through the generation of loss-of-function genetic mouse models for the two ubiquitous B55 isoforms (B55α and B55δ), we report that PP2A-B55α and PP2A-B55δ complexes display overlapping roles in controlling the dynamics of proper chromosome individualization and clustering during mitosis. In the absence of PP2A-B55 activity, mitotic cells display increased chromosome individualization in the presence of enhanced phosphorylation and perichromosomal loading of Ki-67. These data provide experimental evidence for a regulatory mechanism by which the balance between kinase and PP2A-B55 phosphatase activity controls the Ki-67-mediated spatial organization of the mass of chromosomes during mitosis.

The “C”: The large Chameleon-Musca-Coalsack cloud

Recent advancements in 3D dust mapping have transformed our understanding of the Milky Way’s local interstellar medium, enabling us to explore its structure in three spatial dimensions for the first time. In this Letter, we use the most recent 3D dust map by Edenhofer et al. to study the well-known Chameleon, Musca, and Coalsack cloud complexes, located about 200 pc from the Sun. We find that these three complexes are not isolated but rather connect to form a surprisingly well-defined half-ring, constituting a single C-shaped cloud with a radius of about 50 pc, a thickness of about 45 pc, and a total mass of about 5 × 104 M⊙, or 9 × 104 M⊙ if including everything in the vicinity of the C-shaped cloud. Despite the absence of an evident feedback source at its center, the dynamics of young stellar clusters associated with the C structure suggest that a single supernova explosion about 4 Myr–10 Myr ago likely shaped this structure. Our findings support a single origin story for these cloud complexes, suggesting that they were formed by feedback-driven gas compression, and offer new insights into the processes that govern the birth of star-forming clouds in feedback-dominated regions, such as the Scorpius-Centaurus association.

Dynamic Clusters Supported Bi-Level Control System With Lower Communication Requirements for a Distribution Network With Distributed Battery and Photovoltaic Systems

Dynamic Clusters Supported Bi-Level Control System With Lower Communication Requirements for a Distribution Network With Distributed Battery and Photovoltaic Systems

Shapes of dark matter haloes with discrete globular cluster dynamics: The example of NGC 5128 (Centaurus A)

Context. Within the Λ cold dark matter (ΛCDM) cosmology, dark matter haloes are expected to deviate from spherical symmetry. The particular shape of a galactic halo reflects the environment and mass assembly history of its host, as well as the nature of dark matter. Constraining halo shapes at large galactocentric distances is challenging because of the low density of luminous tracers. The well-studied massive early-type galaxy NGC 5128, also known as Centaurus A (Cen A), has a large number of radial velocity measurements for globular clusters (GCs) and planetary nebulae (PNe) extending over a vast area of its extended low-surface-brightness stellar halo. Aims. In this work, we aim to determine the deviation from spherical symmetry of the dark matter halo of Cen A at 5 Re using its GCs as kinematic tracers of the gravitational potential. Methods. We investigated the largest photometric catalogue of GC candidates in order to accurately characterise the spatial distribution of the relaxed population of GCs. To investigate the presence of non-relaxed structures in the kinematic catalogue of GCs, we used the relaxed point-symmetric velocity field as determined by the host’s PNe population. We used anisotropic Jeans modelling under axisymmetric assumptions together with the Gaussian likelihood and GCs as discrete tracers. The gravitational potential is generated by flattened stellar and dark matter distributions. We leveraged the different orbital properties of the blue and red GCs – such as rotation and velocity anisotropy – to model both populations separately. By minimising χ2, we iteratively find the best-fit parameters. Results. We find that the discrete kinematics of the GCs are consistent with being drawn from an underlying relaxed velocity field determined from PNe. The best-fit parameters of the gravitational potential recovered from the blue and red GCs separately agree well and we use them to compute the final results: M200 = 1.86−0.691.61 × 1012 M⊙, M⋆/LB = 2.98−0.78+0.96, and the flattening qDM = 1.45−0.53+0.78. Both GC populations show mild rotation, with red having a slightly stronger rotational signature and radially biased orbits, and blue GCs preferring negative velocity anisotropy. Conclusions. An oblate or a spherical dark matter halo of NGC 5128 is strongly disfavoured by our modelling.

Dynamics of star clusters with tangentially anisotropic velocity distribution

Recent high-precision observations with HST and Gaia enabled new investigations of the internal kinematics of star clusters (SCs) and the dependence of kinematic properties on the stellar mass. These studies raised new questions about the dynamical evolution of self-gravitating stellar systems. We aim to develop a more complete theoretical understanding of how the various kinematical properties of stars affect the global dynamical development of their host SCs. We perform $N$-body simulations of globular clusters with isotropic, radially anisotropic, and tangentially anisotropic initial velocity distributions. We also study the effect of an external Galactic tidal field. We obtain three main results. First, compared to the conventional, isotropic case, the relaxation processes are accelerated in the tangentially anisotropic models and, in agreement with our previous investigations, are slower in the radially anisotropic ones. This leads to, for example, more rapid mass segregation in the central regions of the tangential models or their earlier core collapse. Second, although all SCs become isotropic in the inner regions after several relaxation times, we observe differences in the anisotropy profile evolution in the outer cluster regions --- all tidally filling models gain tangential anisotropy there, while the underfilling models become radially anisotropic. Third, we observe different rates of evolution towards energy equipartition (EEP). While all SCs evolve towards EEP in their inner regions (regardless of the filling factor), the outer regions of the tangentially anisotropic and isotropic models are evolving to an `inverted' EEP (i.e.\ with the high-mass stars having higher velocity dispersion than the low-mass ones). The extent (both spatial and temporal) of this inversion can be attributed to the initial velocity anisotropy --- it grows with increasing tangential anisotropy and decreases as the radial anisotropy rises.

Single and interval valued neutrosophic group decision making problem based on dynamic programming cluster model

Single and interval valued neutrosophic group decision making problem based on dynamic programming cluster model

Wireless optimization for sensor networks using IoT-based clustering and routing algorithms.

Wireless sensor networks (WSN) are among the most prominent current technologies. Its popularity has skyrocketed because of its capacity to operate in difficult situations. The WSN market encompasses various industries, including building automation, security networks, healthcare systems, logistics, and military operations. Therefore, increasing the energy efficiency of these networks is of utmost importance. Hierarchical topology, which typically uses a clustering methodology, is one of the most well-known methods for WSN energy optimization. To achieve energy efficiency in WSN, hierarchical topology low-energy adaptive clustering hierarchy (LEACH) was first introduced, and this served as the foundation. However, conventional LEACH has several limitations, which have led to extensive research into improving LEACH's efficacy in its current form. The use of particular algorithms and strategies to enhance the functionality of the conventional LEACH protocol forms the basis of ongoing efforts. Utilizing this enhanced LEACH, performance in terms of throughput and network life may be enhanced by concentrating on elements such as cluster head formation and transmission energy consumption. The enhanced LEACH algorithm demonstrates significant improvements in both throughput and network lifetime compared with conventional LEACH. Through rigorous experimentation, it was found that the enhanced algorithm increases the throughput by 25% on average, which is attributed to its dynamic clustering and optimized routing strategies. Furthermore, the network lifetime is extended by approximately 30%, primarily because of enhanced energy efficiency through adaptive clustering and transmission power control.

Load aware active low energy adaptive clustering hierarchy for IoT-WSN

Clustering is a primary process that takes place in an IoT based wireless sensor network environment commences from the deployment phase. Due to the heterogeneity and resource constrained nature of internet of things (IoT) networks, dynamic clustering, cluster head selection, and routing are required to optimize the network and to improve the overall network performance. Load aware active low energy adaptive clustering hierarchy (LAALEACH) work is an attempt to introduce novel components to the standard LEACH protocol. The main objective of LAALEACH work is to achieve a load aware active routing in IoT based wireless sensor network environments. Rapid load estimator, load pattern tracker, and load aware active routing are the contributed modules introduced in this LAALEACH work. Most recent related works are analyzed and the proposed modules are devised in a way to overcome the issues in the existing methods. Standard network performance parameters such as throughput, packet delivery rate, communication delays, and energy consumption are measured by the OPNET based simulation during the experiments. Obtained improvements in the overall performance is the accomplishment of LAALEACH work.

Research on mobile robot path planning in complex environment based on DRQN algorithm

A deep reinforcement Q learning algorithm (DRQN) based on radial neural network is proposed to achieve path planning and obstacle avoidance for mobile robots in complex ground environments with different types of obstacles, including static and dynamic obstacles. Firstly, the path planning problem is represented as a partially-observed Markov decision process. Steering angle, running characteristics, and other elements are introduced into the state-action decision space and the greedy factor is dynamically adjusted using a simulated annealing algorithm, which improves the mobile robot’s environment exploration and action selection accuracy. Secondly, the Q-learning algorithm is improved by replacing the Q-table structure with an RBF neural network to enhance the approximation ability of the algorithm’s function values, and the parameters of the implicit layer and the weights between the implicit and the output layer are trained using the dynamic clustering and least-mean methods respectively, which improves the convergence speed and enhances the ability of mobile robots to handle large-scale computation. Lastly, the double reward mechanism is set up to prevent the mobile robot from blind searching in unknown environments, which enhances the learning ability and improves path planning safety and flexibility at the same time. Different types of scenarios are set up for simulation experiments, and the results verified the superiority of the DQRN algorithm. Taking the 30 * 30 complex scene as an example, using the DQRN algorithm for path planning reduces the values of distance, turning angle, and planning time by 27.04%, 7.76%, and 28.05%, respectively, compared to the average values of Q-learning, optimized Q-learning, deep Q-learning, and DDPG algorithms, which can effectively improve the path planning efficiency for mobile robots in complex environments.

A cloud-oriented data-analysis framework to analyze peak demand dynamics in institutional building clusters

Peak loads in higher education institutional building clusters (IBCs) possess considerable economic repercussions on their overall operations. Thus, identifying electrically inefficient buildings presents a significant opportunity to curtail peak loads and promote energy efficiency in IBCs. Existing literature implements clustering algorithms to comprehend the electrical demand dynamics of buildings to analyze disparity in their load behavior. These techniques perform well in single building environment, however, fall short in comprehending the demand dynamics for building clusters, specifically during peak loads. This study introduces a cloud-oriented quantile-based data analysis framework, specifically designed for simultaneously evaluating demand profiles of multiple buildings within IBCs. Quantile-based metrics namely, Value-at-Risk, Conditional Value-at-Risk and Conditional Value-at-Risk standard deviation are implemented to comprehend the electrical fluctuations and quantify the electrical impact of each building during campus-wide peak demands. A cross-building comparison is established by linearly ranking buildings following two key criteria: (i) buildings with frequent demand fluctuations and (ii) buildings exerting a high electrical impact on overall IBC demand. Both criteria are equally weighted while ranking to identify the most inefficient buildings during peak loads. The framework is implemented in a Canadian university and resulted a substantial 50 MW demand reduction through recommissioning and retrofitting of inefficient buildings.