Search

Efficient computational algorithms for bootstrapping linear regression models with clustered data are discussed. For ordinary least squares (OLS) regression, a new algorithm is provided for the pairs cluster bootstrap, along with two algorithms for the wild cluster bootstrap. One of these is a new way to express an existing method. For instrumental variables (IV) regression, an efficient algorithm is provided for the wild restricted efficient cluster (WREC) bootstrap. All computations are based on matrices and vectors that contain sums of squares and cross-products for the observations within each cluster, which have to be computed just once before the bootstrap loop begins. Monte Carlo experiments are used to study the finite-sample properties of bootstrap Wald tests for OLS regression and of WREC bootstrap tests for IV regression.

Density-Based clustering are the main clustering algorithms because they can cluster data with different shapes and densities, but some of these algorithms have high time complexity like OPTICS (Ordering Points to Identify Clustering Structure) and DBSCAN (Density-based spatial clustering of applications with noise) where the time complexity up to O( n 2 ). In this paper, we use an approach to reduce this complexity by providing fuzzy clusters to OPTICS, which make the process of finding neighbours within an only narrow region (fuzzy group) instead of searching all the state space making the algorithm faster than the original one with keeping almost the same accuracy of the innovative algorithm. The results show that there is an improvement in execution time using some synthetic and real datasets with high dimensions.

Dockless E-scooter Sharing (DES) has become a popular means of last-mile commute for many smart cities. As e-scooters are getting deployed dynamically and exibly across city regions that expand and/or shrink, accurate prediction of the e-scooter distribution given the recongured regions becomes essential for city planning. We present GCScoot, a novel ow distribution prediction approach for reconguring urban DES systems. Based on real-world datasets with reconguration, we analyze e-scooter distribution features and ow dynamics for the data-driven designs. We propose a novel spatio-temporal graph capsule neural network within GCScoot to predict future dockless e-scooter ows given the recongured regions. GCScoot pre-processes historical spatial e-scooter distributions into ow graph structures, where discretized city regions are considered as nodes and inter-region ows as edges. To facilitate initial training, we cluster the regions and generate virtual data for new deployment regions based on their peers in the same cluster. Given above designs, the region-to-region correlations embedded within the temporal ow graphs are captured via the multi-graph capsule convolutional neural network which accurately predicts the DES ows. Extensive studies upon four e-scooter datasets (total >3.4 million rides) in four populous US cities have corroborated accuracy and effectiveness of GCScoot in predicting the e-scooter distributions.

So-called spatial web queries retrieve web content representing points of interest, such that the points of interest have descriptions that are relevant to query keywords and are located close to a query location. Two broad categories of such queries exist. The first encompasses queries that retrieve single spatial web objects that each satisfy the query arguments. Most proposals belong to this category. The second category, to which this paper's proposal belongs, encompasses queries that support exploratory user behavior and retrieve sets of objects that represent regions of space that may be of interest to the user. Specifically, the paper proposes a new type of query, the top-k spatial textual cluster retrieval ( $k$ -STC) query that returns the top-k clusters that (i) are located close to a query location, (ii) contain objects that are relevant with regard to given query keywords, and (iii) have an object density that exceeds a given threshold. To compute this query, we propose a DBSCAN-based approach and an OPTICS-based approach that rely on on-line density-based clustering and that exploit early stop conditions. Empirical studies on real data sets offer evidence that the paper's proposals can find good quality clusters and are capable of excellent performance.

In order to solve the problem of hot spot caused by uneven energy consumption of nodes in Wireless Sensor Networks (WSNs) and reduce the network energy consumption, a novel cluster routing algorithm called CRPL for ring based wireless sensor networks using Particle Swarm Optimization (PSO) and Lion Swarm Optimization (LSO) is proposed in this paper. In CRPL, the optimal cluster head (CH) of each ring are selected by using LSO whose fitness function is composed of energy,number of neighbor nodes, number of cluster heads and distance. Moreover, PSO with a multi-objective fitness function considering distance, energy and cluster size is used to find the next hop relay node in the process of data transmission, and the optimal routing paths are obtained, so as to alleviate the hot spot problem as well as decrease the energy consumption in the routing process. The simulation results show that, compared with some existing optimization algorithms, CRPL has better effects in balancing the energy consumption of the network and prolonging the life cycle of the network.

In this paper, we describe our efforts to reproduce the results reported in the SC19 paper by Hidayetoglu et al., titledMemXCT: Memory-Centric X-ray CT Reconstruction withMassive Parallelization. MemXCTs single-device performance, parallelized via OpenMP and MPI using AMD Zen2 CPU cores (second-generation EPYC processor) and NVIDIA V100 GPU devices. Both compute families are available as Microsoft Azurecloud HPC offerings. With these resources, we were are able to reproduce most of the results and exceed the performance of larger inputs on an AMD EPYC HBv2 cluster. We are also able to reproduce the trends of Strong Scaling (SS) for Optimized CPU and GPU versions. Differences in performance of the CPU version were observed due to differences in the underlying hardware, input size, and number of available nodes.

One of the main capabilities of atom probe tomography (APT) is the ability to not only identify but also characterize early stages of precipitation at length scales that are not achievable by other techniques. One of the most popular methods to identify nanoscale clustering in APT data, based on the density-based spatial clustering of applications with noise (DBSCAN), is used extensively in many branches of research. However, it is common that not all of the steps leading to the selection of certain parameters used in the analysis are reported. Without knowing the rationale behind parameter selection, it may be difficult to compare cluster parameters obtained by different researchers. In this work, a simple open-source tool, PosgenPy, is used to justify cluster search parameter selection via providing a systematic sweep through parameter values with multiple randomizations to minimize a false-positive cluster ratio. The tool is applied to several different microstructures: a simulated material system and two experimental datasets from a low-alloy steel . The analyses show how values for the various parameters can be selected to ensure that the calculated cluster number density and cluster composition are accurate.

Previous research on the genesis of industrial clusters has focused on macrolevel (e.g., agglomeration economies and institutions) or mesolevel explanatory factors (e.g., serial entrepreneurship, spin-offs). Less studied are the microfoundations of cluster genesis, intended as the individual- and group-level processes underlying such macrolevel outcomes. Yet, microfoundations are key to understanding the “primordial soup” of cluster genesis—that is, the processes unfolding in the early moments of cluster formation, before the first emergence of commercial activity. Through a historical case study of the British Motorsport Valley (1911–1970s), we trace back the primordial origins of this cluster to the casual leisure activities of groups of amateur motorsport enthusiasts who then prompted the professionalization of motorsport racing and its transformation into the business at the core of the industrial cluster. We theorize that clusters emerge through the layering of different domains (casual leisure, serious leisure, and business), each made of three elements (actors, activities, and artifacts), which interact via two microlevel mechanisms: (1) localizing passion, a shared emotional energy by which people become affectively attached to the spaces where they carry out activities that they enjoy; (2) domain repurposing, the shift of a configuration of actors, activities, and artifacts toward a new purpose, originating a new domain. Whereas domain repurposing induces the transformation of activities from leisure to business (thus originating the industry at the core of a cluster), localizing passion anchors the activities to the same geographical area (clustering the industry). Our key contribution is to explore the emotional microfoundations of cluster genesis.

In this article, we investigate the distributed Nash equilibrium (NE) seeking problem for the multi-cluster game under switching communication topologies. Specifically, the communication topology switches between a group of jointly connected digraphs. First, a new distributed NE seeking algorithm for the multi-cluster games is designed by the consensus protocol and gradient play rule under the switching communication topologies. Furthermore, in order to make the algorithm still applicable when the agent only knows part of the decision information, the leader-following consensus protocol is used to generate the estimates for all agents action in the cluster under the assumption that the switching topology between clusters is directed and strongly connected. A more general NE seeking algorithm for the multi-cluster games is designed. For these two algorithms, the results of local convergence and non-local convergence are given, respectively. Two examples verify the validity of the theoretical results.

In this paper, we investigate temporal clusters of extremes defined as subsequent exceedances of high thresholds in a stationary time series. Two meaningful features of these clusters are the probability distribution of the cluster size and the ordinal patterns giving the relative positions of the data points within a cluster. Since these patterns take only the ordinal structure of consecutive data points into account, the method is robust under monotone transformations and measurement errors. We verify the existence of the corresponding limit distributions in the framework of regularly varying time series, develop non-parametric estimators and show their asymptotic normality under appropriate mixing conditions. The performance of the estimators is demonstrated in a simulated example and a real data application to discharge data of the river Rhine., Deutsche Forschungsgemeinschaft, Projekt DEAL

For the first time, the present work shows a combined theoretical and experimental analysis of the preferred adsorption sites for both methanol and methoxy species over Cu, Pt, and bimetallic Cu-Pt clusters - up to 5 atoms in size - supported on CeO2(111) as model systems. DFT+U calculations compared to methanol adsorption experiments followed by DRIFT spectroscopy over Cu, Pt, and bimetallic Cu-Pt nanoparticles supported over CeO2 octahedra, exposing only {111} crystal planes. From our DFT+U calculations, we found that methanol adsorption over the CuPt/CeO2(111) system is slightly favored, compared to the pristine CeO2(111) surface. Similarly, methoxy species formed through methanol dehydrogenation (CH3O- + H+) adsorb preferably over the metallic clusters, compared to the pristine oxide surface. This promotion is explained by the higher coordination of the methoxy species over the cluster, especially for those containing a higher amount of Cu atoms. This results highlights the oxidized Cu sites' role in the catalytic process, followed by the surface cerium atoms reduction of the support. The interaction of methanol with the substrate oxygen weakens the O-H bond, which results in a reduction in the energy barrier when the reaction occurs on the surface and interface, compared when this occurs exclusively on the metal cluster. However, after this step, the potential energy surface's depth suggests that methoxy species prefer to be adsorbed over the metal clusters. Experimentally, methanol was adsorbed over Cu, Pt, and bimetallic Cu-Pt nanoparticles supported over CeO2 {111}-octahedral particles. Through in-situ DRIFTS, bands related to adsorbed methanol and methoxy species were identified. A blue shift in all the bands was observed, related to metals’ influence in adsorbed species' vibrational mode. In the experimental spectra, new contributions in Cu-containing catalysts appeared, which were identified as adsorbed species over the metallic cluster and the interface, rationalized by calculated spectra.

In order to explore novel high-energy and green oxidants, oxygen-rich clusters O4−/0 and O6−/0 were investigated by laser-vaporization source, mass spectrometry and theoretical calculations. Most of the low-lying isomers of On−/0 clusters are polyline structures consisting of O2 and O3 units. With one exception, O6, which has a chair-form cyclic structure with D3d symmetry, just like cyclohexane. It has the largest HOMO-LUMO energy gap of the On−/0 (n = 4, 6) clusters. The energy performances of O6 cluster were also estimated, which the density is 2.03 g/cm3, the solid enthalpy of formation is 114.75 kcal/mol, and the theoretical specific impulse (Isp) of H2(L)/O6 cluster is calculated to be 507.78 s.

C2H2 semi-hydrogenation has been widely applied in industry to eliminate trace C2H2 from C2H4 feed. C2H2 semi-hydrogenation to C2H4 on a series of the newly designed catalysts, graphdiyne (GDY) as a new carbon allotrope supported different sizes of PdxMy clusters (PdxMy/GDY, M = Cu, Ag, Au, Ni; x+y = 1–3), were studied using DFT calculations. The results found that C2H2 semi-hydrogenation to C2H4 on PdxMy/GDY catalysts exhibits that both the activity and selectivity greatly depend on the composition and size of PdxMy/GDY catalysts. Surprisingly, our results for the first time discovered the Pd1/GDY catalyst with GDY supported the single atom Pd that presents the best selectivity and activity toward C2H4 formation compared to the previously reported catalysts so far in C2H2 semi-hydrogenation. This study would provide a theoretical clue for designing and screening out the potential catalysts with GDY supported small sizes of PdxMy and other metal clusters in C2H2 hydrogenation.

The person re-identification (ReID) method in a single-domain achieves appealing performance, but its reliance on label information greatly limits its extensibility. Therefore, the unsupervised cross-domain ReID method has received extensive attention. Its purpose is to optimize the model by using the labelled source domain and the unlabelled target domain and finally make the model well generalized in the target domain. We propose an unsupervised cross-domain ReID method based on median stable clustering (MSC) and global distance classification (GDC). Specifically, the measurement method used by MSC comprehensively considers the similarity between clusters, the number of samples in a cluster, and the combined similarity within a cluster. Different from the method based on triple loss, GDC can separate the distance distribution of positive and negative sample pairs in a global scope. In addition, considering that model performance is very sensitive to probability parameters when source domain memory is reconsolidated, we designed a dynamic memory reconsolidation (DMR) method to reduce the influence of parameters on performance. Extensive experiments on large-scale datasets (Market-1501, DukeMTMC-reID and MSMT17) demonstrate the superior performance of MSC-GDC over the state-of-the-art methods.

Energy efficiency is considered as the most potential issue in Wireless Sensor Networks (WSNs), since they incorporate limited sized batteries that could not be recharged or replaced. The energy possessed by the sensor nodes needs to be optimally utilized in order to extend the lifetime expectancy with guaranteed QoS in the network. In this paper, a Hybrid Artificial Bee Colony and Monarchy Butterfly Optimization Algorithm (HABC-MBOA)-based Cluster Head Selection Scheme is proposed for the predominant selection of cluster heads under clustering process. This proposed HABC-MBOA replaces the employee bee phase of ABC with mutated butterfly adjusting operator of MBOA for preventing earlier trapping of solutions into a local optimal point and delayed convergence by maintaining the tradeoff between exploitation and exploration. This proposed HABC-MBOA plays an anchor role in eliminating inadequacy of ABC algorithm towards global search potential. This proposed HABC-MBOA also eliminates the possibility of cluster heads being overloaded with maximum number of sensor nodes, that results in rapid death of the sensor nodes during the deployment of impotent cluster head selection process. The simulation results confirmed that the number of alive nodes in the network is determined to be 18.92% superior to the benchmarked cluster head selection approaches.

[EN] Porous Fe carboxylates are amongst the most promising MOF-based materials due to their low price, low toxicity, metal environments (including open metal sites) and remarkable (meso)porosity variety. Fe-MOFs based on the cluster [FeO(X)(solvent)], that is, MIL-101(Fe), MIL-100(Fe) and semiamorphous Fe-BTC, are of particular interest. These three materials are quite related each other: (i) MIL-100(Fe) and MIL-101(Fe) have the same zeolitic topology MTN and two types of mesocavities, whereas (ii) Fe-BTC and MIL-100(Fe) form an unprecedented semiamorphous / fully-crystallized pair, having in common the metal cluster, the composition, one mesocavity, etc but without becoming a nano- / micro-crystalline pair. This work describes the room-temperature synthesis, characterization and catalytic performance in the aerobic cyclohexene oxidation of the semiamorphous Fe-BDC, which together with MIL-101(Fe) would form the second semiamorphous / crystallized pair in MOFs. Unfortunately, Fe-BDC could not be prepared in water as solvent, but in either ethanol or in N,N-dimethylformamide. It possesses relatively high textural properties (above 500 mg) and key common features with MIL-101(Fe): XRD reflections at the same 2θ positions, similar thermal stability, almost equal linker conformations, etc. Fe-BDC became quite active in the solvent-free aerobic oxidation of cyclohexene under mild conditions, surpassing the activity performance of the well-known commercial Fe-BTC catalyst in the same reaction under the same mild conditions., This work has been partially financed by CSIC through the programs: (i) CVCSIC-AEPP-Ayudas Extraordinarias para preparación de proyectos 2019 (2019AEP076) and (ii) i-COOP-2018 (COOPA20271). This work has been also financed by the CONACyT project A1–5-30646. J.G.F. acknowledges a Ph.D. CONACyT grant (687839). R.D.G. thanks the JAE-Intro CSIC grant (JAEINT18_EX_0043).

This technical note investigates the cluster consensus control for general linear multi-agent systems. The focus of this technical note is especially to solve the following problem: in what kinds of connection scenarios can agents achieve cluster consensus regardless of the coupling strength (CCRCS). Different from existing literature, this technical note extends the case to general partition graphs, including both acyclic and cyclic partition cases. New analysis methods are put forward. It is proved that, under some basic assumptions, the multi-agent systems can achieve CCRCS if and only if the interaction topology satisfies the non-vicious circle condition, which is proposed for the first time. An algorithm is presented which can check whether a given interaction topology satisfies the non-vicious circle condition. The result can also be extended to heterogeneous cases that agents from different clusters possess non-identical system dynamics.

A cluster-based drag model is proposed for the gas–solid circulating fluidized bed (CFB) riser by including the cluster information collected from image processing and wavelet analysis into the calculation of system drag. The performance of the proposed drag model is compared with some commonly used drag models. A good agreement with the experimental data is achieved by the proposed cluster-based drag model. Error analysis of the proposed cluster-based drag model based on the local distributions of solids holdup and particle velocity is conducted. The clustering phenomenon in the low-density and high-density CFB risers and the effect of the cluster size on the simulation accuracy are also numerically studied by the proposed drag model.

Battery energy storage system (BESS) plays an important role in the grid-scale application due to its fast response and flexible adjustment. Energy loss and inconsistency of the battery will degrade the operating efficiency of BESS in the process of power allocation. BESS usually consists of many energy storage units, which are made up of parallel battery clusters with a cell-pack-cluster hierarchical structure. This paper presents a power allocation strategy based on cluster switching to relieve the stated problem in two levels. Cluster switching is identified as a new control approach to eliminating the imbalanced state-of-charge (SOC) in the cluster level. In the unit level, an optimization model is constructed for power allocation, where the objective function consists of two aspects: minimizing battery energy loss and maximizing SOC consistency. Finally, the effectiveness of the power allocation strategy for BESS in peak shaving and frequency regulation is validated by simulation and scaled-down experiment. The results prove that the power allocation strategy can reduce the battery energy loss and prevent from overcharging/overdischarging to extend the battery lifetime.

Purpose This paper aims to evaluate whether the integration of smart specialization strategies (S3) into clusters significantly impacts their efficiency for countries that still do not implement this policy. This study tests three effects: whether the kind of policies envisaged through an S3 strategy impacts cluster’s efficiency; whether this impact changes with the technological intensity of the clusters; to determine which S3 is more suitable for sub-clusters at different levels of technological intensity. Design/methodology/approach The Mexican economy is taken as case of study because it has a proper classification of its industries intro Porter’s cluster’s definition but still does not adopt the S3 policy. Through data envelopment analysis (DEA), this study evaluates the cluster’s efficiency increment when variables representing the S3 elements are included. Findings The results show that strategies following the S3 had a significant impact in all clusters, but when clusters were classified by technological intensity, the impact on efficiency is higher in clusters in the medium low-tech group. Practical implications According to the results in the DEA, it can be concluded that these S3 strategies have the potential to increase the clusters’ productivity significantly. These results make convenient the adoption of the S3 policy by countries that already count with a properly cluster definition. Originality/value These findings contribute to the lack of studies that analyze the join implementation of S3 on clusters.

I present a critical overview of so-called "{\it ab initio}" DFT (density fuctional theory) based calculation schemes for the description of the electronic structure, energy spectrum, and optical response for strongly correlated 3$d$ oxides, in particular, crystal-field and charge transfer transitions as compared with an "old"\, cluster model that does generalize crystal-field and ligand-field theory. As a most instructive illustration of validity of numerous calculation techniques I address the prototypical 3$d$ insulator NiO predicted to be a metal in frames of a standard LDA (local density approximation) band theory., Comment: 21 pages, 1 figure

Hidayetolu et al. (2019) proposed a novel memory-centric computation system, MemXCT. As a challenge at SC20, we reproduce the computational efficiency of MemXCT on our Azure cloud cluster. Our experiments evaluate the overall performance and the strong scalability with real datasets and verify part of the conclusions in the original paper. Digital artifacts from these experiments are available at: 10.5281/zenodo.5598108.

In this brief, we investigate the cluster synchronization of multi-weighted complex networks with pinning control. We assume that the network contains multiple coupling matrices, and they can be all asymmetric (including the symmetric case), which is more practical in the real world. For multiple asymmetric coupling matrices, the Lyapunov function is hard to design. We firstly define a candidate region for every matrix. Then if these regions are not too far, we can find (at least) a common vector lying in each region. This method has geometrical meaning, so it is different from previous LMI method. That is to say, we can design a new Lyapunov function with a vector combining all normalized left eigenvectors (NLEVec) corresponding to the zero eigenvalue of coupling matrices. With some sufficient conditions presented accordingly, the realization of the cluster synchronization is guaranteed. Especially, we carefully discuss the case with three coupling matrices. The adaptive rule for cluster synchronization is also designed. Finally, numerical simulations are given to demonstrate our results’ effectiveness.

This paper establishes asymptotic results for the maximum likelihood and restricted maximum likelihood (REML) estimators of the parameters in the nested error regression model for clustered data when both of the number of independent clusters and the cluster sizes (the number of observations in each cluster) go to infinity. Under very mild conditions, the estimators are shown to be asymptotically normal with an elegantly structured covariance matrix. There are no restrictions on the rate at which the cluster size tends to infinity but it turns out that we need to treat within cluster parameters (i.e. coefficients of unit-level covariates that vary within clusters and the within cluster variance) differently from between cluster parameters (i.e. coefficients of cluster-level covariates that are constant within clusters and the between cluster variance) because they require different normalisations and are asymptotically independent.

Wind farm cluster penetration and its long‐distance delivery have become the important form of wind power accommodation of power network in China. For the adaptability of power network planning to the output volatility of clustering wind farms, three steps are studied as follows. Initially, the output uncertainty model of clustering wind farms based on Auto Regression Integrated Moving Average (ARIMA)‐Generalized Autoregression Conditional Heteroscedasticity (GARCH)‐Pair Copula is put forward to analyze the intermittency, fluctuation and clustering wind power correlation. Secondly, Full Cost (FC) theory is introduced to account social expense brought by grid‐connected clustering wind farms from four aspects of static cost, dynamic cost, environmental cost and inter‐generational cost. Meanwhile, available wind power transmission capacity is defined to reflect wind power accommodation based on transmission margin. Finally, the adaptive efficiency index is proposed by the ratio of available wind power transmission capacity to FC. The transmission adaptability planning model considering cluster correlation of wind farms is built with the objective function of the adaptive efficiency index. Taking the actual regional power grid as an example, the results show that the planning scheme considering the optimal adaptive efficiency index of clustering wind farms can be better adapted to the future power network environment.

Partitioning data into internally homogeneous parts is an important problem when mining in situ engineering data. In this paper, a polynomial regression-based fuzzy c-means (PR-FCM) clustering algorithm that utilizes the functional relationships among the attributes of the input dataset is proposed. In this algorithm, a polynomial regression equation is taken as the center of each cluster instead of the cluster prototype used in conventional FCM, and the difference between a sample and a cluster prototype is defined as the distance between the actual value of one attribute and the corresponding predicted value provided by its own polynomial regression equation. An alternating optimization method is designed to optimize the new clustering objective function of the proposed algorithm. A series of experiments on synthetic and real-world datasets are conducted to evaluate the performance of the PR-FCM algorithm, which exhibits higher effectiveness and possesses more advantages than the original FCM algorithm. The PR-FCM algorithm is applied to tunnel boring machine (TBM) operation data from a TBM project in China. The experimental results show that the proposed algorithm can effectively cluster TBM operation data.

Perovskite and organic solar cells usually require electron-transport interlayers to efficiently transport electrons from the photoactive layer to the metal electrode. In general, pure organic or i...

Atmospheric organic acids (OAs) are expected to enhance methanesulfonic acid (MSA)-driven new particle formation (NPF). However, the exact role of OAs in MSA-driven NPF remains unclear. Here, we employed a two-step strategy to probe the role of OAs in MSA–methylamine (MA) NPF. Initially, we evaluated the enhancing potential of 12 commonly detected OAs in ternary MA–MSA–OA cluster formation by considering the formation free energies of the (MSA)1(MA)1(OA)1 clusters and the atmospheric concentrations of the OAs. It was found that formic acid (ForA) has the highest potential to stabilize the MA–MSA clusters. The high enhancing potential of ForA results from its acidity, structural factors such as no intramolecular H bonds, and high atmospheric abundance. The second step is to extend the MSA–MA–ForA system to larger cluster sizes. The results indicate that ForA can indeed enhance MSA–MA NPF at atmospheric conditions (the upper limited temperature is 258.15 K), indicating that ForA might have an important role in MSA-driven NPF. The enhancing effect of ForA is mainly caused by an increased formation of the (MSA)2(MA)1 cluster, which is involved in the pathway of binary MSA–MA nucleation. Hence, our results indicate that OAs might be required to facilitate MSA-driven NPF in the atmosphere.

One often observes small but measurable differences in the diffraction data measured from different crystals of a single protein. These differences might reflect structural differences in the protein and may reveal the natural dynamism of the molecule in solution. Partitioning these mixed-state data into single-state clusters is a critical step that could extract information about the dynamic behavior of proteins from hundreds or thousands of single-crystal data sets. Mixed-state data can be obtained deliberately (through intentional perturbation) or inadvertently (while attempting to measure highly redundant single-crystal data). To the extent that different states adopt different molecular structures, one expects to observe differences in the crystals; each of the polystates will create a polymorph of the crystals. After mixed-state diffraction data have been measured, deliberately or inadvertently, the challenge is to sort the data into clusters that may represent relevant biological polystates. Here, this problem is addressed using a simple multi-factor clustering approach that classifies each data set using independent observables, thereby assigning each data set to the correct location in conformational space. This procedure is illustrated using two independent observables, unit-cell parameters and intensities, to cluster mixed-state data from chymotrypsinogen (ChTg) crystals. It is observed that the data populate an arc of the reaction trajectory as ChTg is converted into chymotrypsin.

Subsequence clustering is an important issue in time series data mining. Observing that most time series consist of various patterns with different unknown lengths, we propose an optimization framework to adaptively estimate the lengths and representations for different patterns. Our framework minimizes the inner subsequence cluster errors with respect to subsequence clusters and segmentation under time series cover constraint where the subsequence cluster lengths can be variable. To optimize our framework, we first generate abundant initial subsequence clusters with different lengths. Then, three cluster operations, i.e., cluster splitting, combination and removing, are used to iteratively refine the cluster lengths and representations by respectively splitting clusters consisting of different patterns, joining neighboring clusters belonging to the same pattern and removing clusters to the predefined cluster number. During each cluster refinement, we employ an efficient algorithm to alternatively optimize subsequence clusters and segmentation based on dynamic programming. Our method can automatically and efficiently extract the unknown variable-length subsequence clusters in the time series. Comparative results with the state-of-the-art are conducted on various synthetic and real time series, and quantitative and qualitative performances demonstrate the effectiveness of our method.

Recently, many real-world applications exploit multi-view data, which is collected from diverse domains or obtained from various feature extractors and reflect different properties or distributions of the data. In this work, a novel unsupervised multi-view framework is proposed to cluster such data. The proposed method, called Multi-View clustering with Adaptive Sparse Memberships and Weight Allocation (MVASM), pays more attention to constructing a common membership matrix with proper sparseness over different views and learns the centroid matrix and its corresponding weight of each view. Concretely, MVASM method attempts to learn a common and flexible sparse membership matrix to indicate the clustering, which explores the underlying consensus information of multiple views, and solves the multiple centroid matrices and weights to utilize the view-specific information and further modifies the above-mentioned membership matrix. In addition, the theoretical analysis, including the determination of the power exponent parameter, convergence analysis, and complexity analysis are also presented. Compared to the state-of-the-art methods, the proposed method improves the performance of clustering on different public datasets and demonstrates its reasonability and superiority.

Three-body $AAB$ model for the $NN{\bar K}(s_{NN}=0)$ kaonic cluster is considered based on the configuration space Faddeev equations. Within a single-channel approach, the difference between masses of nucleons and kaons and the charge independence breaking of nucleon-nucleon interaction are taken into consideration. We definite the particle configurations in the system according to the particle masses and pair potentials. There are two sets of the particle configurations, $ ppK^-$, $np \bar{K^0}$ and $nn {\bar K}^0$, $npK^-$, charged and neutral. The three-body calculations are performed by applying $NN$ and $N\bar K$ phenomenological isospin-dependent potentials. The mass and energy spectra related to the particle configurations are presented. We evaluate the mass and energy uncertainties for the $NN\bar K$ model. An analogy to $NNN$ model for the $^3$H and $^3$He nuclei is proposed.

Prediction of redox potentials is essential for catalysis and energy storage. Although density functional theory (DFT) calculations have enabled rapid redox potential predictions for numerous compounds, prominent errors persist compared to experimental measurements. In this work, we develop machine learning (ML) models to reduce the errors of redox potential calculations in both implicit and explicit solvent models. Training and testing of the ML correction models are based on the diverse ROP313 dataset with experimental redox potentials measured for organic and organometallic compounds in a variety of solvents. For the implicit solvent approach, our ML models can reduce both the systematic bias and the number of outliers. ML corrected redox potentials also demonstrate less sensitivity to DFT functional choice. For the explicit solvent approach, we significantly reduce the computational costs by embedding the microsolvated cluster in implicit bulk solvent, obtaining converged redox potential results with a smaller solvation shell. This combined implicit-explicit solvent model, together with GPU-accelerated quantum chemistry methods, enabled rapid generation of a large dataset of explicit-solvent-calculated redox potentials for 165 organic compounds, allowing detailed investigation of the error sources in explicit solvent redox potential calculations.

This paper proposes a cluster-aware supervised learning (CluSL) framework, which integrates the clustering analysis with supervised learning. The objective of CluSL is to simultaneously find the best clusters of the data points and minimize the sum of loss functions within each cluster. This framework has many potential applications in healthcare, operations management, manufacturing, and so on. Because CluSL, in general, is nonconvex, we develop a regularized alternating minimization (RAM) algorithm to solve it, where at each iteration, we penalize the distance between the current clustering solution and the one from the previous iteration. By choosing a proper penalty function, we show that each iteration of the RAM algorithm can be computed efficiently. We further prove that the proposed RAM algorithm will always converge to a stationary point within a finite number of iterations. This is the first known convergence result in cluster-aware learning literature. Furthermore, we extend CluSL to the high-dimensional data sets, termed the F-CluSL framework. In F-CluSL, we cluster features and minimize loss function at the same time. Similarly, to solve F-CluSL, a variant of the RAM algorithm (i.e., F-RAM) is developed and proven to be convergent to an [Formula: see text]-stationary point. Our numerical studies demonstrate that the proposed CluSL and F-CluSL can outperform the existing ones such as random forests and support vector classification, both in the interpretability of learning results and in prediction accuracy. Summary of Contribution: Aligned with the mission and scope of the INFORMS Journal on Computing, this paper proposes a cluster-aware supervised learning (CluSL) framework, which integrates the clustering analysis with supervised learning. Because CluSL is, in general, nonconvex, a regularized alternating projection algorithm is developed to solve it and is proven to always find a stationary solution. We further generalize the framework to the high-dimensional data set, F-CluSL. Our numerical studies demonstrate that the proposed CluSL and F-CluSL can deliver more interpretable learning results and outperform the existing ones such as random forests and support vector classification in computational time and prediction accuracy.

Time-of-flight secondary ion mass spectra measured for different carbon allotropes under Bi3+ cluster ion bombardment were acquired. Ion yield of homogeneous (Cn)- and heterogeneous metal-carbon (CsCn)- secondary cluster ions was examined. Qualitative explanation of the regularities was given based on the concept of nonlinear atomic collision cascades. It was suggested that that the features of secondary cluster yield can be used for determination of hybridization state of the target material. Unlike most other techniques, secondary ion mass spectrometry is capable of the sputter depth profiling of the hybridization state.

Dimensionality Reduction (DR) techniques can generate 2D projections and enable visual exploration of cluster structures of high-dimensional datasets. However, different DR techniques would yield various patterns, which significantly affect the performance of visual cluster analysis tasks. We present the results of a user study that investigates the influence of different DR techniques on visual cluster analysis. Our study focuses on the most concerned property types, namely the linearity and locality, and evaluates twelve representative DR techniques that cover the concerned properties. Four controlled experiments were conducted to evaluate how the DR techniques facilitate the tasks of 1) cluster identification, 2) membership identification, 3) distance comparison, and 4) density comparison, respectively. We also evaluated users' subjective preference of the DR techniques regarding the quality of projected clusters. The results show that: 1) Non-linear and Local techniques are preferred in cluster identification and membership identification; 2) Linear techniques perform better than non-linear techniques in density comparison; 3) UMAP (Uniform Manifold Approximation and Projection) and t-SNE (t-Distributed Stochastic Neighbor Embedding) perform the best in cluster identification and membership identification; 4) NMF (Nonnegative Matrix Factorization) has competitive performance in distance comparison; 5) t-SNLE (t-Distributed Stochastic Neighbor Linear Embedding) has competitive performance in density comparison., Comment: IEEE VIS 2021, to appear in IEEE Transactions on Visualization & Computer Graphics

Understanding driver interaction behavioral semantics has potential benefits to autonomous car's decision-making design. This article presents a framework of analyzing various encountering behaviors through decomposing driving encounter sequential data into small building blocks, called traffic primitives, using a Bayesian nonparametric learning (BNPL) approach. This framework offers a flexible way to gain semantic insights into complex driving encounters without any prerequisite knowledge of interaction behavior categories. Its effectiveness is then validated using 976 naturalistic driving encounters from which more than 4000 traffic primitives were learned with the BNPL approach. After that, a dynamic time warping method integrated with k-means clustering is then developed to cluster all these extracted traffic primitives into groups. Experimental results identify 20 kinds of traffic primitives capable of representing the essential components of driving encounters in our database. Based on the results, we conclude that the proposed primitive-based analysis could prove useful for autonomous vehicle applications.

Hypothesis: While the lack of efficient tools yielding controllable uniform saturation ratios (S) has delayed basic experimental heterogeneous nucleation studies, common diffusive condensation particle counters (DCPCs) could fill this gap if their S-variation were minimized by increasing the proportion of sheath gas (σ) surrounding a central core of purified clusters. Analysis: We measure the activation probability P of Tetraheptylammonium Bromide cluster cations (THA-Br)n-1THA+ in Kanomax’s fast CPC while controlling S through the saturator and condenser temperatures (Ts, Tc), varying σ, and changing the size (n) of purified salt clusters via high resolution mobility selection. Findings: Experimental curves P(Ts,n) obtained in 1-butanol/air at fixed Tc (13 °C) and variable n and Ts (3 ≤ n ≤ 16; 30 ≤ Ts ≤ 40 °C) rise sharply versus both n and Ts. Their steepness increases five-fold with increasing σ to about σ = 75%, with little effect thereafter. Measurements changing S would yield size distributions of unknown aerosols at fairly high resolution. Comparing P(Ts,n) data with predictions from capillary theory suggests that basic heterogeneous nucleation measurements can be carried out, but instrument improvements are still needed.

The charge distributions and elastic electron form factors for 18O, 42,44Ca, 58Ni, and 118Sn nuclei were considered using the cluster expansion of the 1- and 2-body terms. Inelastic electron form factors to 2+ states with core-polarization effects were studied, where the nuclear collective modes were considered alongside the shell model transition density. The influence of short-range correlations (SRCs) was calculated by the parameter β, which was introduced into the ground-state charge distribution and concluded the Jastrow-type function. It is found that the inclusion of 2-body correlations is necessary to describe the observed data of elastic and inelastic form factors at high momentum transfer q > 3 fm–1.

Here, we conducted a photometric and astrometric study of two open stellar clusters Dolidze-18 and Ruprecht-70, which have not been photometrically studied before. The most important thing for using Gaia (DR2) database lies in the positions, parallax, and proper motions, which make us split cluster members from the field ones and get precise astrophysical parameters. On studying the radial density profiles of these clusters, the actual sizes are estimated and found larger using Gaia. From the color-magnitude diagrams and theoretical isochrones, we simultaneously determined the ages, distance moduli, and reddening of the two clusters. However, considering the parallaxes of Gaia (DR2) for the cluster members, we calculated the cluster distance and confirmed what we obtained from the color-magnitude diagram. Then, the Cartesian galactocentric coordinates (Xo, Yo, Zo), and the distances from the galactic center (Rg) were also estimated. According to the luminosity and mass functions, the total luminosity and total mass of the clusters are estimated. Our study shows that Ruprecht-70 is recently dynamically relaxed, while Dolidze 18 is not relaxed yet., Comment: 17 pages, 8 figures, 1 table

Let G = S L ( 2 , 5 ) be the special linear group of 2 × 2 -matrices with coefficients in the field with 5 elements. We show that the principal block over a splitting field K of characteristic two of the group algebra KG has a 3-cluster tilting module. This gives the first example of a representation-infinite block of a group algebra having a cluster tilting module and answers a question by Erdmann and Holm.

The purpose of this study is to obtain the results of the analysis of factors that make prospective customers who want to open an account at a branch switch to opening an online account. This study uses the K-Means method and uses the age factor, the time period in opening an account as a parameter. The data collection technique is in the form of data collection by questionnaire. The subject of this research is marketing who are members of PT Dika in collaboration with PT Bank Central Asia. Based on the research conducted by the author using the K-Means Clustering method and the rapidminer application, the researchers drew the following conclusions; 1) Of the three clusters, the age factor that has the most value is at age 21 in cluster 3 as many as 3 people. As for the age of 22 to 26 in the three clusters, the three clusters are not much different, 2) Of the three clusters, the most income factors are in group 3 as many as 4 people in cluster 2, namely earning around 1.500.000 – 3,000,000, 3) Of the three cluster, the distance factor from home to the nearest BCA Bank is at most at distance 1 as many as 3 people in cluster 3 and at distance 3 as many as 3 people in cluster 2, 4) Of the three clusters, the processing time factor is mostly in group 1 as much as 3 people in cluster 1, group 2 as many as 3 people in cluster 3 and group 3 as many as 3 people in cluster 3, and 5) From the three clusters, the factor of moving from opening an account at a branch to opening an online account (pemol) is because it's faster, in cluster 1 as many as 3 people, in cluster 2 as many as 2 people and in cluster 3 as many as 3 people.

Novel research avenues have been explored over the last decade on the use of transition metal carbides (TMCs) as catalytically active supports for metal nanoclusters, which display high catalytic activity despite the poor reactivity (or even inertness) of the bulk metal. It has been postulated that TMCs polarise the electron density of adsorbed metal particles in such a way that their catalytic activity ends up being superior to those dispersed on more traditional metal oxide supports. Herein, we investigate the structural and electronic properties of small clusters of precious metals (Rh, Pd, Pt and Au) and more affordable metals (Co, Ni and Cu) supported on TMCs with 1:1 stoichiometry (TiC, ZrC, HfC, VC, NbC, TaC, MoC and WC) by means of periodic Density Functional Theory calculations. Our high-throughput screening studies indicate that it is possible not only to have strongly bonded and stably dispersed metal nanoparticles on TMC surfaces, but also to manipulate their charge by carefully selecting elements with desired electronegativity for the host TMC and the metal cluster. By doing so, it is possible to tune the amount of charge density on the cluster hollow sites, which can facilitate the bonding of certain molecules. Moreover, we identify Pt, Pd and Rh clusters supported on hexagonal TMC (001) facets as the candidates with the highest potential catalytic activity -as estimated by the significant polarisation of the cluster electron density- and stability -as quantified by the strongly negative values of adsorption energy per atom and formation energy-.

Multibaseline (MB) phase unwrapping (PU) is a vital processing procedure for MB synthetic aperture radar interferometry (InSAR) signal processing and can improve the traditional InSAR by changing the ill-posed problem to the well-posed problem. The existing research has shown that the MB PU problem can be successfully converted into an unsupervised cluster analysis problem. Using the high feature descriptiveness of the deep learning technique, an unsupervised deep convolutional neural network, referred to as CANet, is proposed to cluster all the pixels into different groups according to the input's recognizable pattern of the ambiguity number of the MB interferometric phase. Subsequently, we extend our previous two-stage programming-based MB processing approach (TSPA) to processing MB PU on a sparse irregular network, which is established from the clustering result of CANet. Both theoretical analysis and experimental results show that the proposed method is an effective MB PU method, and its execution time is drastically lower than those of many classical MB PU methods.

We present new algorithms to classify structural ensembles of macromolecules, based on the recently proposed extended similarity measures. Molecular Dynamics provides a wealth of structural information on systems of biologically interest. As computer power increases we capture larger ensembles and larger conformational transitions between states. Typically, structural clustering provides the statistical mechanics treatment of the system to identify relevant biological states. The key advantage of our approach is that the newly introduced extended similiarity indices reduce the computational complexity of assessing the similarity of a set of structures from O(N2) to O(N). Here we take advantage of this favorable cost to develop several highly efficient techniques, including a linear-scaling algorithm to determine the medoid of a set (which we effectively use to select the most representative structure of a cluster). Moreover, we use our extended similarity indices as a linkage criterion in a novel hierarchical agglomerative clustering algorithm. We apply these new metrics to analyze the ensembles of several systems of biological interest such as folding and binding of macromolecules (peptide,protein,DNA -protein). In particular, we design a new workflow that is capable of identifying the most important conformations contributing to the protein folding process. We show excellent performance in the resulting clusters (surpassing traditional linkage criteria), along with faster performance and an efficient cost-function to identify when to merge clusters.