Dynamic Clustering

Clustered sensor networks have recently been shown to increase system throughput, decrease system delay, and save energy while performing data aggregation. Whereas those with rotating cluster heads, such as LEACH (low-energy adaptive clustering hierarchy), have also advantages in terms of security, the dynamic nature of their communication makes most existing security solutions inadequate for them. In this paper, we investigate the problem of adding security to hierarchical (cluster-based) sensor networks where clusters are formed dynamically and periodically, such as LEACH. For this purpose, we show how random key predistribution, widely studied in the context of flat networks, and mTESLA, a building block from SPINS, can be both used to secure communications in this type of network. We present our solution, and provide a detailed analysis of how different values for the various parameters in such a system impact a hierarchical network in terms of security and energy efficiency. To the best of our knowledge, ours is the first that investigates security in hierarchical WSNs with dynamic cluster formation.

Recent combined experiments by small angle neutron scattering (SANS) and neutron spin echo (NSE) have demonstrated that dynamic clusters can form in concentrated lysozyme solutions when the right combination of a short-ranged attraction and a long-ranged electrostatic repulsion exists. In this paper, we investigate the temperature effect on the dynamic cluster formation and try to pinpoint the transition concentration from a monomeric protein phase to a cluster phase. Interestingly, even at a relatively high concentration (10% mass fraction), despite the significant change in the SANS patterns that are associated with the change of the short-ranged attraction among proteins, the normalized short-time self-diffusion coefficient is not affected between 5 and 40 • C. This is interpreted as a lack of cluster formation in this condition. However, at larger concentrations such as 17.5% and 22.5% mass fraction, we show that the average hydrodynamic radius increases significantly and causes a large decrease of the normalized self-diffusion coefficient as a result of cluster formation when the temperature is changed from 25 to 5 • C.

Clustering is an important research topic in wireless networks, because cluster structures can facilitate resource reuse and increase system capacity. Ad hoc networks consist of wireless hosts that communicate with each other in the absence of a fixed infrastructure. Such networks cannot rely on centralized and organized network management. A slight change in the speed of cluster head nodes has a great influence on the cluster members and even causes the cluster head to switch frequently. we present a robust self-stabilizing weight-based clustering algorithm for ad hoc networks. The robustness property guarantees that, starting from an arbitrary configuration, after one asynchronous round, the network is partitioned into clusters. The communication in VANET is at greater risk because the messages are broadcasted and a wireless medium is used for data propagation. For secure communication among these vehicles, clustering algorithms are preferred. The proposed algorithm intends to create stable clusters by reducing reclustering overhead prolonging cluster lifetime, and shortening the average distance between cluster heads and their cluster members. Many clustering schemes have been proposed by the researchers which can be classified into either mobility based clustering or non-mobility based clustering algorithms. This technique takes the distance and velocity as a parameter to create relatively a stable cluster structure. We also developed a algorithm for super cluster-head selections. .

The paper analyzes the relationship between innovation dimensions, according to the European Innovation Scoreboard, and the entrepreneurship rate in a group of moderate innovator countries. Using the Pearson linear correlation, the author conducted a data analysis based on an empirical study using open data from the Summary Innovation Index -European Innovation Scoreboard (SII-EIS) and Eurostat for 2013-2019. This period covers the moment when all the countries of the Visegrad Group (i.e., the Czech Republic, Hungary, Poland, and Slovakia -the V4) were classified into the same innovation group according to EIS. However, due to the volume limitations of the article, it was decided that Moderate Innovators from 2019 would be included in the comparative analysis, i.e., when the V4 were last collectively classified as Moderate Innovators. The results of the research positively verify the initial research hypothesis that the EIS indicators show a different strength of relationships with the entrepreneurship rate in the V4 and among other economies classified as Moderate Innovators (according to SII-2020). The variables that describe the quantity and quality of entrepreneurs' innovations strongly and positively impact the V4 countries' entrepreneurship rate. In other countries, the variables derived from the investment attractiveness of economies show a significant and positive correlation with the entrepreneurship rate. Based on the analysis of the results, it can be concluded that there is a strong positive correlation between the entrepreneurship rate and the total innovation index, which is compiled each year based on a set of variables for the European Union countries. The study showed that the entrepreneurship rate in these countries is strongly impacted by indicators representing the following groups: Innovators (small and medium-sized enterprises (SMEs) with product or process innovations; SMEs with marketing or organizational innovations,

Data clustering is an approach for automatically finding classes, concepts, or groups of patterns. It also aims at representing large datasets by a few number of prototypes or clusters. It brings simplicity in modelling data and plays an important role in the process of knowledge discovery and data mining. Data mining tasks require fast and accurate partitioning of huge datasets, which may come with a variety of attributes or features. This imposes computational requirements on the clustering techniques. Swarm Intelligence (SI) has emerged that meets these requirements and has successfully been applied to a number of real world clustering problems. This paper looks into the use of Particle Swarm Optimization for cluster analysis. The effectiveness of Fuzzy C-means clustering provides enhanced performance and maintains more diversity in the swarm and allows the particles to be robust to trace the changing environment. Data structure identifying from the large scale data has become a ...

Different numerical mappings of the DNA sequences have been studied using a new cluster-scaling method and the well known spectral methods. It is shown, in particular, that the nucleotide sequences in DNA molecules have robust cluster-scaling properties. These properties are relevant to both types of nucleotide pair-bases interactions: hydrogen bonds and stacking interactions. It is shown that taking into account the cluster-scaling properties can help to improve heterogeneous models of the DNA dynamics. It is also shown that a chaotic (deterministic) order, rather than a stochastic randomness, controls the energy minima positions of the stacking interactions in the DNA sequences on large scales. The chaotic order results in a large-scale chaotic coherence between the two complimentary DNA-duplex's sequences. A competition between this broad-band chaotic coherence and the resonance coherence produced by genetic code has been briefly discussed. The Arabidopsis plant genome (which is a model plant for genome analysis) and two human genes: BRCA2 and NRXN1, have been considered as examples.

Scalability presents a significant challenge in vehicular communication, particularly when there is no hierarchical structure in place to manage the increasing number of vehicles. As the number of vehicles increases, they may encounter the broadcast storm problem, which can cause network congestion and reduce communication efficiency. Clustering can solve these issues, but due to high vehicle mobility, clustering in vehicular ad hoc networks (VANET) suffers from stability issues. Most of the existing clustering protocols are optimized for highways only and some protocols are optimized for intersections only. The clusters that are created by the protocols which are optimized for highways breakdown at the intersections where clustering is needed more. On the other hand, the clusters that are created by the protocols optimized for intersections create extra clusters or break unnecessary on straight-roads which reduces cluster stability. Moreover, the lack of intelligent use of a combination of the mobility parameters, such as direction, movement, position, velocity, degree of vehicle, and movement at the intersection, also contributes to cluster stability problems. A dynamic clustering protocol that efficiently combines all the mobility parameters can resolve these issues in VANETs by providing stability in both intersections and highways. To achieve high stability in VANET clustering, a Stable Dynamic feedback-based Predictive Clustering (SDPC) protocol is proposed for VANET to ensure cluster stability in both highway and intersection scenarios. SDPC considers vehicle relative velocity, acceleration, position, distance, transmission range, moving direction at the intersection, and vehicle density to create a cluster. The cluster head is selected based on the future position of the vehicles, relative distance between vehicles, movement of vehicles at the intersection, degree of vehicles, and probable cluster head duration. The performance of SDPC is compared with four existing VANET clustering protocols in various road topologies, in terms of the average cluster head change rate, the average duration of cluster head, the average duration of cluster member, and the average clustering overhead. The simulation results show that SDPC outperforms the existing protocols, achieving higher clustering stability in terms of cluster head change rate (50%), cluster head duration (15%), cluster member duration (6%), and clustering overhead (35%).

Core operations (e.g. TDMA scheduler, synchronized sleep period, data aggregation) of many proposed protocols for different layer of sensor network necessitate clock synchronization. Our paper mingles the scheme of dynamic clustering and diffusion based asynchronous averaging algorithm for clock synchronization in sensor network. Our proposed algorithm takes the advantage of dynamic clustering and then applies asynchronous averaging algorithm for synchronization to reduce number of rounds and operations required for converging time which in turn save energy significantly than energy required in diffusion based asynchronous averaging algorithm.

Generating schedules such that all operations are repeated every constant period of time is as important as generating schedules with minimum delays in all cases where a known discipline is desired or obligated by stakeholders. In this paper, a periodic job shop scheduling problem (PJSSP) based on the periodic event scheduling problem (PESP) is presented, which deviates from the cyclic scheduling. The PESP schedules a number of recurring events as such that each pair of event fulfills certain constraints during a given fixed time period. To solve such a hard PJSS problem, we propose a hybrid algorithm, namely PSO-SA, based on particle swarm optimization (PSO) and simulated annealing (SA) algorithms. To evaluate this proposed PSO-SA, we carry out some randomly constructed instances by which the related results are compared with the proposed SA and PSO algorithms as well as a branch-and-bound algorithm. In addition, we compare the results with a hybrid algorithm embedded with electromagnetic-like mechanism and SA. Moreover, three lower bounds (LBs) are studied, and the gap between the found LBs and the best found solutions are reported. The outcomes prove that the proposed hybrid algorithm is an efficient and effective tool to solve the PJSSP.

This paper presents an approach to improving the scalability of the decentralized smart contract platform Waterfall, based on the concept of hierarchical fractal sharding. Although distributed ledger technology holds significant promise for building secure and transparent digital ecosystems, its widespread adoption remains limited by scalability issues. A key challenge lies in the inability to proportionally increase transaction throughput with the growing number of participants without undermining either decentralization or security. The proposed solution reduces both computational and communication loads by distributing transactions, smart contracts, and network state across a system of recursively structured shards. Each shard operates as an independently validated subnetwork organized as a directed acyclic graph structure that supports asynchronous execution and consensus. This design enables the participation of low-power nodes, enhances load balancing, and achieves scalability not only at the level of the entire network but also within its internal components. The study details the mechanisms for shard formation and merging, transaction routing strategies, and dynamic placement of smart contracts. In addition, a probabilistic model is introduced to evaluate the risk of malicious capture of individual shards, and guidelines are provided for choosing safe shard sizes under various threat assumptions. While the proposed architecture is designed specifically for the Waterfall platform, its core principles and several of its methods may be adapted to other distributed ledger systems, including but not limited to blockchain-based platforms, particularly those employing modular or directed acyclic graph-structured architectures.

The Self-Organizing Map is well-known as the unsupervised classification method. It is employed as classifier in various applications such as image segmentation. The main purpose of this paper is to identify and detect an object of interest on side scan sonar image. This work is performed by two steps. The first one is to split an image into regions of uniform texture using the Gray Level Co-occurrence Matrix Method (GLCM) which is widely used in texture segmentation application. The last one address the unsupervised learning method based on the Artificial Neural Networks (Self-Organizing Map or SOM) used for determining the comparative model of object of interest from an image. To increase the performance of SOM, we propose a penalty function based on data histogram visualization. After a brief review of both techniques (GLCM and SOM), we present our method and some results from several experiments on the real world data set.

Reactions between 6Li and 6Li were studied in the laboratory energy range of 2-16 MeV. Angular distributions and excitation functions were measured for various proton, deuteron, triton and e-channels. In addition excitation functions were obtained for five 7transitions in ~~ and the ground-state transitions in both 7Li and VBe. Detailed statistical model calculations were performed under various assumptions concerning the formation of the compound nucleus. It is concluded that a smooth cutoff of the entrance channel angular momenta yields the only consistent description for both angular distributions and excitation functions. Only a few proton channels are dominated by compound nucleus decay. All other channels reflect the complex structure of 6Li which causes a variety of direct reaction modes. Nuclear Reactions: 6Li + 6Li at 2 < Ela b ~ 16 MeV; measured angular distributions and excitation functions of p, d, t, e-channels and y-transitions in 7Li, 7Be, l~ statistical model calculations; deduced fusion cross section; reaction mechanism inferred.

How to process the big data generated in large IoT networks is still challenging current techniques. To date, a lot of network clustering approaches have been proposed to improve the performance of data aggregation in IoT. However, most of them focus on partitioning networks with static topologies, and thus they are not optimal on handling the case with moving objects in the networks. Moreover, as the best of knowledge, none of them has ever considered the performance of following computing in edge servers. To improve these problems, in this work, we propose a highly efficient IoT network dynamic clustering solution in edge computing using deep reinforcement learning (DRL). Our approach can both fulfill the data communication requirements from IoT networks and load-balancing requirements from edge servers, and thus provide a great opportunity for future high performance IoT data analytics. We implement our approach by Deep Q-learning Network (DQN) model, and our preliminary experimental results show that the DQN solution can achieve higher score in cluster partitioning compared with the current static benchmark solution.

Background: This paper addresses key biological problems and statistical issues in the analysis of large gene expression data sets that describe systemic temporal response cascades to therapeutic doses in multiple tissues such as liver, skeletal muscle, and kidney from the same animals. Affymetrix time course gene expression data U34A are obtained from three different tissues including kidney, liver and muscle. Our goal is not only to find the concordance of gene in different tissues, identify the common differentially expressed genes over time and also examine the reproducibility of the findings by integrating the results through meta analysis from multiple tissues in order to gain a significant increase in the power of detecting differentially expressed genes over time and to find the differential differences of three tissues responding to the drug. Bayesian categorical model for estimating the proportion of the 'call' are used for pre-screening genes. Hierarchical Bayesian Mixture Model is further developed for the identifications of differentially expressed genes across time and dynamic clusters. Deviance information criterion is applied to determine the number of components for model comparisons and selections. Bayesian mixture model produces the gene-specific posterior probability of differential/non-differential expression and the 95% credible interval, which is the basis for our further Bayesian meta-inference. Meta-analysis is performed in order to identify commonly expressed genes from multiple tissues that may serve as ideal targets for novel treatment strategies and to integrate the results across separate studies. We have found the common expressed genes in the three tissues. However, the up/down/no regulations of these common genes are different at different time points. Moreover, the most differentially expressed genes were found in the liver, then in kidney, and then in muscle. Despite rapid advancements in statistical methods for gene expression microarray analysis, much more work is needed for multiple source heterogeneous genomic data, such as multiple organisms/tissues, multiple platforms, multiple species and even more from transcriptome, genome, to proteome in order to develop valid and dependable methods that are mainly applicable to microarray data. The congruency of these different data sources needs a unified framework for combining the multiple

Unit commitment (UC) programming is a critical task in power system operations, which faces problems such as uncertainty in generation and loads with the significant rise in the generation of electrical energy through renewable energy sources (RES) such as wind and responsive load programs. The problem of UC, or the unit commissioning problem, is a major optimization problem, the exact solution of which can lead to a significant reduction in costs. In this article, smart grids are considered which aim to reduce costs and environmental problems. Thus, this paper solves the UC problem in smart grids by considering the emission of generation units, resulting in a multi-objective function for minimization. With the introduction of smart grids, energy storage systems (ESS) have also been considered in the grid. This paper proposes the optimal charging and discharging of ESS. Another problem modeled in this article is that of demand response (DR) in smart grids. To validate the performance of the proposed model, it is tested on a 4-unit system with ESS and the results show its optimal performance. To solve the problem of UC programming, a hybrid honey bee mating and bacterial foraging algorithm are used to reduce the complexity of the problem and achieve optimal results.

Host–parasitoid systems with evolving mutation rates are studied. By increasing the growth rate of hosts, the diversity of both species is maintained dynamically. For the lower growth rate, diversity is brought about by mere parasitism. The average mutation rate for parasites is elevated to a high value, while that for hosts is suppressed at a low level. For the higher growth rate, the mutation rates for both hosts and parasites are elevated to form a symbiotic cluster connected by on-going mutation. This symbiotic state is sustained through a chaotic oscillation keeping some coherency among species. For a flat landscape for hosts, dynamical clustering of oscillation is observed. Lyapunov spectra of such oscillations show that high dimensional chaos with small positive exponents underlies in the symbiotic state. This weak high dimensional chaos, termed ‘‘homeochaos,’’ is essential to the maintenance of symbiosis in ecosystems.

In this paper we present a new algorithm for document clustering called Condensed Star (ACONS). This algorithm is a natural evolution of the Star algorithm proposed by Aslam et al., and improved by them and other researchers. In this method, we introduced a new concept of star allowing a different star-shaped form; in this way we retain the strengths of previous algorithms as well as address previous shortcomings. The evaluation experiments on standard document collections show that the proposed algorithm outperforms previously defined methods and obtains a smaller number of clusters. Since the ACONS algorithm is relatively simple to implement and is also efficient, we advocate its use for tasks that require clustering, such as information organization, browsing, topic tracking, and new topic detection.

PurposeThe purpose of this paper is to explore the factors influencing entrepreneurial intensity. More specifically the study addresses the following objectives: propose a way to measure entrepreneurial intensity within the community context in order to determine entrepreneurial activity over a period of ten years, detect the factors influencing the entrepreneurial intensity, and finally locate Kibbutz communities on the entrepreneurial grid.Design/methodology/approachKibbutz communities are the level of analysis. Using a comprehensive questionnaire, a sample of 60 Kibbutzim – constituting 22 percent of the population of Kibbutz communities in Israel – was investigated over a period of ten years. The same questionnaire was administered to the same sample Kibbutzim (Kibbutzim is the plural of Kibbutz) in 1994, 1997 and 2004. Collected data include number and types of enterprises, economic strength, organizational size and age, and features of organizational structure and culture.Find...

Interoperability plays an important role for a variety of applications. One of them are Peer Data Management Systems, where autonomous data sources (peers) interact with each other based on semantic mappings between their schemas. The building blocks that enable interoperability and thus the main challenges in such systems are mapping representation, query rewriting, and efficient query processing. While most approaches regard these aspects in separate this paper presents a comprehensive study of the interactions between these blocks. Our considerations try to provide a holistic view on semantic interoperability in distributed environments such as PDMS. We discuss techniques for distributed query processing and rewriting that consider high-level query operators such as top-N and skyline. Furthermore, we discuss how to increase efficiency by applying routing indexes and relaxation of result completeness/correctness.

Contemporary computers collect databases that can be too large for classical methods to handle. The present work takes data whose observations are distribution functions (rather than the single numerical point value of classical data) and presents the computational statistical approach of a new methodology to group the distributions into classes. The clustering method links the searched partition to the decomposition of mixture densities, through the notions of a function of distributions and of multidimensional copulas. The new clustering technique is illustrated by ascertaining distinct temperature and humidity regions for a global climate dataset and shows that the results compare favorably with those obtained from the standard EM algorithm method.

It is shown that the nucleotide sequences in DNA molecules have clusterscaling properties (discovered for the first time in turbulent processes: Sreenivasan and Bershadskii, 2006, J. Stat. Phys., 125, 1141-1153.). These properties are relevant to both types of nucleotide pair-bases interactions: hydrogen bonds and stacking interactions. It is shown that taking into account the cluster-scaling properties can help to improve heterogeneous models of the DNA dynamics. Two human genes: BRCA2 and NRXN1, have been considered as examples.

In wireless sensor networks, failures occur due to energy depletion, environmental hazards, hardware failure, communication link errors, etc. These failures could prevent them to accomplish their tasks. Moreover, most routing protocols are designed for ideal environment such as LEACH. Hence, if nodes fail the performance of these protocols degrade. In this context, we propose two improved versions of LEACH so that it becomes a fault-tolerant protocol. In the first version, we propose a clustered architecture for LEACH in which there are two cluster-heads in each cluster: one is primary (CHp) and the other is secondary (CHs). In the second version, we propose to use the checkpoint technique. Finally, we conducted several simulations to illustrate the performance our contribution and compared obtained results to LEACH protocol in a realistic environment.

News clustering, categorization and analysis are key components of any news portal. They require algorithms capable of dealing with dynamic data to cluster, interpret and to temporally aggregate news articles. These three tasks are often solved separately. In this paper we present a unified framework to group incoming news articles into temporary but tightly-focused storylines, to identify prevalent topics and key entities within these stories, and to reveal the temporal structure of stories as they evolve. We achieve this by building a hybrid clustering and topic model. To deal with the available wealth of data we build an efficient parallel inference algorithm by sequential Monte Carlo estimation. Time and memory costs are nearly constant in the length of the history, and the approach scales to hundreds of thousands of documents. We demonstrate the efficiency and accuracy on the publicly available TDT dataset and data of a major internet news site.

Dynamical mean-field embedding of the dual fermion dynamical cluster approach for strongly correlated systems ZI YANG MENG, SANDEEP PATHAK, SHUXIANG YANG, HANNA TERLETSKA, JUANA MORENO, MARK JARRELL, Louisiana State University -We extend the recently developed dual fermion dynamical cluster approach with a further embedding of the dual fermion lattice into a larger, third length scale. The resulting approach is a complete multi-scale many-body technique for strongly correlated electron systems. It treats the short length scales explicitly by the dynamical cluster approach, intermediate length scales diagrammatically with the dual fermion technique, and the largest length scales approximately at a dynamical mean-field level. This technique iterated to self-consistency on all the three length scales. To illustrate the implementation and applicability of this method, we test it with the one and two dimensional Falicov-Kimball model. We will specifically address the convergence and critical scaling behavior of the charge-density-wave transition temperature.

This paper discusses processes of technological change in the tile cluster in the village Karanggeneng in Central Java, Indonesia. A growing number of producers in this cluster have switched from traditional kiln to so-called handpress production. We will analyze the processes of innovation adoption of the handpress technology in the cluster. Adoption of the handpress technology encompasses certain indivisibilities which require collaboration among producers to render innovation adoption profitable. We will discuss how pioneer adopters share information and stimulate social learning and adoption on the part of their neighbours, because simultaneous adoption of the handpress technology by groups of producers is needed in order to tackle the indivisibilities which pioneers are facing.

This paper gives purpose to identify the factors of the constitution of "the Tunisian cluster" in an environment open to competition. It is a contribution to the debate on the importance of clusters for competitiveness of small and medium enterprises (SMEs) to make more innovative and competitive regions and to promote strategically important sectors in technology. Approaches based on the knowledge economy grew by better integrating forms of proximity, organizational, institutional and geographical, (Torre and Rallet 2005) and relational (Boshma, 2005). An empirical study was conducted on a sample of Information and Communication Technologies ICTs'companies. The results show that the lack of attractiveness of ICT Tunisian companies to form clusters is not due to a lack of suitable infrastructure but to the absence of relations involved in a partnership approach or localized nature between higher education, research centers, industry training and organization, enabling to carry out scientific and technical projects.

The study of possible associations between social and health factors was one of the purposes of the multicentric Italian survey of perinatal preventive medicine. To get a concise and coherent social indicator, which could be useful to this study, we explored the socioeconomic information systematically collected on the parents of 12,058 babies born in 1973-1975 in four centres (Trieste, Milan, Parma and Bari). which reflect some of the heterogeneous aspects of Italy. The Correspondence analysis indicated that the seven considered indexes. concerning both parents' education levels and occupations, dwelling quality and the length of father's residence in the area in which the baby was born. were highly and similarly correlated in each centre. The Dynamic cluster analysis enabled us to bring together the socioeconomic profiles of the families into six classes. on the basis of their similarities, which depend only upon the multivariate distribution of the seven categorical variables under study. Therefore, the attained classification not only is independent of whatever score is assigned to the socioeconomic categories but also allows for the strong interactions between the variables and exploits all the available information. Our six classes were found to be rather similar to the six socioeconomic groups (S.E.G.) of the U.K. Registrar General and to show some resemblance to certain social groups defined by other classifications adopted in obstetric and pediatric field. The socioeconomic indicator constructed here should enable us to find out how much, in different areas of Italy, the social classes may account for inequalities in health conditions and care of the mother and her baby.

Recent spacecraft observations show evidence of the 3-µm water of hydration absorption feature across the lunar surface. We examine Lunmos 9 and Lunmap 14 detections of an absorption feature at 0.7-µm suggesting oxidized iron in phyllosilicates.

In this paper we introduce a new unsupervised segmentation algorithm for textured sonar images. A Dynamic Self-Organizing Maps (DSOM) algorithm capable of incremental learning has been developed to automatically cluster the input data into relevant classes of seabed. DSOM algorithm is an extension of classical Self-Organizing Maps (SOM) algorithm combined with Adaptive Resonance Theory (ART) technique. The proposed approach is based on growing map size during learning processes. Starting with a minimal number of neurons, the map size increases dynamically and the growth is controlled by the vigilance threshold of the ART network. To assess the consistency of the proposed approach, the DSOM algorithm is first tested on simulated data sets and then applied on real sidescan sonar images. The results obtained using the proposed approach demonstrate its capability to successfully cluster sonar images into their relevant seabed classes, very close to those resulting from human expert interpretation.

In this paper we introduce a new unsupervised segmentation algorithm for textured sonar images. A Dynamic Self-Organizing Maps (DSOM) algorithm capable of incremental learning has been developed to automatically cluster the input data into relevant classes of seabed. DSOM algorithm is an extension of classical Self-Organizing Maps (SOM) algorithm combined with Adaptive Resonance Theory (ART) technique. The proposed approach is based on growing map size during learning processes. Starting with a minimal number of neurons, the map size increases dynamically and the growth is controlled by the vigilance threshold of the ART network. To assess the consistency of the proposed approach, the DSOM algorithm is first tested on simulated data sets and then applied on real sidescan sonar images. The results obtained using the proposed approach demonstrate its capability to successfully cluster sonar images into their relevant seabed classes, very close to those resulting from human expert interpretation.

A model is presented that simulates the process of neuronal synchronization, formation of coherent activity clusters and their dynamic reorganization in the olivo-cerebellar system. Three coupled 2D lattices dealing with the main cellular groups in this neuronal circuit are used to model the dynamics of the excitatory feedforward loop linking the inferior olive (IO) neurons to the cerebellar nuclei (CN) via collateral axons that also proceed to terminate as climbing fiber afferents to Purkinje cells (PC). Inhibitory feedback from the CN-lattice fosters decoupling of units in a vicinity of a given IO neuron. It is shown that noise-sustained oscillations in the IO-lattice are capable to synchronize and generate coherent firing clusters in the layer accounting for the excitable collateral axons. The model also provides phase resetting of the oscillations in the IO-lattices with transient silent behavior. It is also shown that the CN-IO feedback leads to transient patterns of couplings in the IO and to a dynamic control of the size of clusters.

The unstable heavier nuclei release energy spontaneously by emitting radiations and particles, the process called radioactivity, namely, alpha, beta and gamma radioactive decays. In addition to these three types of decays, there exists fourth type of radioactivity known as cluster radioactivity (CR), defined as the spontaneous emission of fragments or clusters (A ≥ 4) heavier than alpha particle but less than a fission fragment. The decays of number of radioactive nuclei in trans-lead region, which lead to doubly magic Pb (Z=82, N=126) or neighboring daughter nuclei through emission of various clusters have been explored in many experimental as well as theoretical studies during last 3-4 decades. The shell closure in the atomic nucleus has now been established as the prominent cause of the process of CR.

This work is devoted to to the research into the possibilities of improving Proof-of-Work blockchain technology based on dynamic clustering of nodes to reduce transaction time.To eliminate the problem with the Proof-of-Work mechanism, which is considered in this work, it is necessary to solve the problem of reducing the amount of time spent on a transaction. It is proposed to implement this by dividing the system into subnets: when the consensus is not accepted by the entire community, but it is accepted by groups separately-thus minimizing the transaction time in the Proof-of-State algorithm. There is no ready-made solution for the Proof-of-Work dynamic consensus mechanism that would be successfully applied in blockchain technologies. All existing algorithms for dividing the blockchain network into subgroups are used only for static algorithms, but Proof-of-Work is dynamic and has certain features: there is no scope; the user does not see the list of nodes. These features greatly complicate the implementation of clustering for the Proof-of-Work consensus mechanism. The task of this study is the formulation of hypotheses and the verification of the formulated hypotheses, which are aimed at increasing the speed of the transaction. For verification, it is proposed to simulate a blockchain network to conduct experiments and test hypotheses that can potentially solve the Proof-of-Work problem. To develop a way for improving the Proof-of-Work consensus mechanism by implementing dynamic clustering of network nodes, flexibility and scalability, minimal impact on the existing blockchain protocol are taken into account, security issues are also important, management of the process of accepting new nodes to avoid possible attacks and ensure integrity and network security. Existing methods of improving Proof-of-Work technology, clustering methods that can be applied to the network are analyzed; problems that arise when developing a new technique are identified. A blockchain network modeling system has been developed and implemented, with the help of which the approach of dynamic grouping of nodes of the blockchain network, in which the system is divided into subsystems, is implemented. The results of the study allow us to conclude: the cluster system gives improved values of the number of transactions per second (by two hundredths transactions), average transaction time (by one and sixty-seven hundredthsseconds), throughput (by two tenthstransactions), transaction delay (by one and six hundred sixty-seven thousandths seconds) and significantly reduces the total energy consumption of the system (a difference of five thousand, one hundred twenty-two units). This indicates the potential of the proposed method in various practical applications.

The FORUM project aims at extending existing data integration techniques in order to facilitate the development of mediation systems in large and dynamic environments. It is well known from the literature that a crucial point that hampers the development and wide adoption of mediation systems lies in the high entry and maintenance costs of such systems. To overcome these barriers,

In Libya, from time to time, the National Electricity Grid is directed by the National Electricity Company to conduct load shedding to alleviate pressure on supply at times of peak demand. This involves hours’ of power outages in the area covered by this study, namely, the Southern Electrical Grid of Libya (SEGL). The proposed approach focuses on a set of region categories to mitigate the effects of the load shedding problem by considering the demand priorities on the operation of the power system during emergencies. The classification of these region categories is solved by using a relatively new algorithm, the harmony search (HS) algorithm, which is applied to a randomly generated list of electricity deficits and region categories to identify those regions that should not experience a power outage at times of peak load. The proposed approach is applied to 50 simulated cases and compared to actual SEGL cases. The obtained results show that the proposed algorithm has a high level of...

In this paper, we present a novel dynamic clustering algorithm that efficiently deals with data streams and achieves several important properties which are not generally found together in the same algorithm. The dynamic clustering algorithm operates online in two different timescale stages, a fast distancebased stage that generates micro-clusters and a density-based stage that groups the micro-clusters according to their density and generates the final clusters. The algorithm achieves novelty detection and concept drift thanks to a forgetting function that allows micro-clusters and final clusters to appear, drift, merge, split or disappear. The outlier identification is made in a natural way using micro-clusters density. This algorithm has been designed to be able to detect complex patterns even in multi-density distributions and making no assumption of cluster convexity. The performance of the dynamic clustering algorithm is assessed theoretically through complexity analysis and empirically through a set of experiments that compare the algorithm with other algorithms of the literature on popular data sets that have interesting characteristics regarding the emergence, disappearance and movement of clusters as well as multi-density, non-convexity and noise.

This paper presents a mobile agent information system designed as the basic tool for creating dynamic clusters. It is aimed at mobile and heterogeneous environments where needs for complex resources and services may arise from stationary or mobile devices of different computing capabilities. MAIS is part of a larger project that will offer users necessary tools for using in a simple manner aggregated heterogeneous and mobile resources and services. Experimental evaluations, reported in this paper, prove the effectiveness and flexibility of the MAIS system.

Considering the Indian context as the basis of enquiry into industrial clustering three important issues that largely determine the performance and strategy of firms to survive compete and grow are identified. These three issues relate to the characteristics of the existing and potential ...

Context. Observational and instrumental difficulties observing small bodies below 0.5 μm make this wavelength range poorly studied compared with the visible and near-infrared. Furthermore, the suitability of many commonly used solar analogues, essential in the computation of asteroid reflectances, is usually assessed only in visible wavelengths, while some of these objects show spectra that are quite different from the spectrum of the Sun at wavelengths below 0.55 μm. Stars HD 28099 (Hyades 64) and HD 186427 (16 Cyg B) are two well-studied solar analogues that instead present spectra that are also very similar to the spectrum of the Sun in the wavelength region between 0.36 and 0.55 μm. Aims. We aim to assess the suitability in the near-ultraviolet (NUV) region of the solar analogues selected by the team responsible for the asteroid reflectance included in Gaia Data Release 3 (DR3) and to suggest a correction (in the form of multiplicative factors) to be applied to the Gaia DR3 aste...

Studies of the thermodynamic properties of nuclear matter require accurate methods of temperature determination. Double-ratios of isotope yields which cancel out chemical potential effects offer a particularly promising technique of temperature determination, known as the Albergo-method [1]. The open and solid circular points in Fig. 1 show the measured isotopic yield ratios, d/t (left panel) and 3 He/ 4 He (middle panel), and the double ratio, R H-He (right panel), for central 112 Sn + 112 Sn and 124 Sn + 124 Sn collisions, respectively. For both reactions, 112 Sn + 112 Sn and 124 Sn + 124 Sn, the measured double ratio is the same, R H-He = 10, within experimental errors. This observation is consistent with the results of ref. [2]. Thus, the isotope double ratio is independent of the isospin of the system as is required for consistency with equilibrium that is necessary for fragmentation models. This value for the R H-He double ratio is obtained from the two isotope ratios, d/t and 3 He/ 4 He, both of which depend, as expected [3], strongly on the proton-to-neutron ratio of the reaction involved. Both the d/t and the 3 He/ 4 He isotope ratios are about 40% greater for the 112 Sn + 112 Sn reaction than for 124 Sn + 124 Sn. As a consequence, the R H-He double ratio has the same value for both systems.

Research in sensor networks has focused on development of energy efficient and secure infrastructures. In this article, we introduce a new approach to organize sensor networks in clusters in order to reduce energy dissipation. Our contribution is an heuristic to define the number of clusters and also an efficient manner to choose cluster heads by minimizing the distance between the cluster heads and its cluster nodes. Inspired from LEACH, a well-known TDMA cluster-based sensor network architecture, we introduce a new method for building and maintaining clusters using the paradigm of a soccer team. In this work, a new algorithm called OH-Kmeans, based on the Kmeans algorithm, is used to find dynamically the number of clusters and form them guaranteeing direct transmission between the cluster heads and cluster nodes.

The coefficients of the successiveh 2 orders of the extended Thomas-Fermi expansion of the kinetic energy density are obtained to the leading order in the density inhomogeneities. This is done by comparing the corresponding static polarizability with the exact value in two cases: a non-interacting nuclear infinite system and symmetric nuclear matter with a Skyrme interaction.

we propose a new perspective on the identification of linear dynamic system using structural similarity. The proposal consists in the meaningful exploration of each model, specifically behavior of the state variable. The decomposition of the behavior of a state variable in different modes of behavior of a system, each one has a different set of weights and shows different patterns of behavior. These weights are more significant than eigenvalue to develop a new technique for identifying linear system and invariants over time. We use two methods based on different areas of knowledge such as linear algebra and statistics. This paper is a conceptual proof that enriches the implementation and validity not only from point of view algorithmic likewise physic mathematical.

This paper addresses dynamic data clustering as an optimization problem and propose techniques for finding optimal (number of) clusters in a multi-dimensional data or feature space. In order to accomplish this objective we first propose two novel techniques, which successfully address several major problems in the field of Particle Swarm Optimization (PSO) and promise a significant breakthrough over complex, multi-modal optimization problems at high dimensions. The first one, so-called Multi-Dimensional (MD) PSO, reforms the native structure of swarm particles in such a way that they can make inter-dimensional passes with a dedicated dimensional PSO process. Therefore, in a multidimensional search space where the optimum dimension is unknown, swarm particles can seek both positional and dimensional optima. This eventually removes the necessity of setting a fixed dimension a priori, which is a common drawback for the family of swarm optimizers. Nevertheless, MD PSO is still susceptible to premature convergences due to lack of divergence. To address this problem we propose Fractional Global Best Formation (FGBF) technique, which basically collects all promising dimensional components and fractionally creates an artificial global-best particle (aGB) that has the potential to be a better "guide" than the PSO's native gbest particle. We investigated both individual and mutual applications of the proposed techniques and demonstrated that the best clustering performance can be achieved by their mutual operation. In order to test and evaluate their clustering performance in terms of accuracy, robustness and scalability, a synthetic data-set, which contains ground-truth clusters and offers a broad range of complexity levels is used. An extensive set of experiments demonstrate that the proposed dynamic clustering technique based on MD PSO with FGBF can extract the optimal (number of) clusters by converging to the global optimum of the validity index function at the true dimension.

In the current model of early Solar System evolution, the stable members of the Jovian and Neptunian Trojan populations were captured into resonance from the leftover reservoir of planetesimals during the outward migration of the giant planets. As a result, both Jovian and Neptunian Trojans share a common origin with the primordial disk population, whose other surviving members constitute today's trans-Neptunian object (TNO) populations. The cold (low inclination and small eccentricity) classical TNOs are ultra-red, while the dynamically excited "hot" (high inclination and larger eccentricity) population of TNOs contains a mixture of ultra-red and blue objects. In contrast, Jovian and Neptunian Trojans are observed to be blue. While the absence of ultra-red Jovian Trojans can be readily explained by the sublimation of volatile material from their surfaces due to the high flux of solar radiation at 5 AU, the lack of ultra-red Neptunian Trojans presents both a puzzle and a challenge to formation models. In this work we report the discovery by the Dark Energy Survey (DES) of two new dynamically stable L4 Neptunian Trojans, 2013 VX 30 and 2014 UU 240 , both with inclinations i > 30 • , making them the highest-inclination known stable Neptunian Trojans. We have measured the colors of these and three other dynamically stable Neptunian Trojans previously observed by DES, and find that 2013 VX 30 is ultra-red, the first such Neptunian Trojan in its class. As such, 2013 VX 30 may be a "missing link" between the Trojan and TNO populations. Using a simulation of the DES TNO detection efficiency, we find that there are 162 ± 73 Trojans with H r < 10 at the L4 Lagrange point of Neptune. Moreover, the blue-to-red Neptunian Trojan population ratio should be higher than 17:1. Based on this result, we discuss the possible origin of the ultra-red Neptunian Trojan population and its implications for the formation history of Neptunian Trojans.

Recent cell biological studies suggest that sphingolipids and cholesterol may cluster in biomembranes to form raft-like microdomains. Such lipid domains are postulated to function as platforms involved in the lateral sorting of certain proteins during their trafficking within cells as well as during signal transduction events. Here, the physical interactions that occur between cholesterol and sphingolipids in model membrane systems are discussed within the context of microdomain formation. A model is presented in which the role of cholesterol is refined compared to earlier models.