Search NASA⌕ Search

SEARCH · Search NASA

Results for “clustering methods”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 649 records · Page 36

Tensor decomposition in post-Hartree–Fock methods. II. CCD implementation

In a previous publication, we have discussed the usage of tensor decomposition in the canonical polyadic (CP) tensor format for electronic structure methods. There, we focused on two-electron integrals and second order Møller-Plesset perturbation theory (MP2). In this work, we discuss the CP format for Coupled Cluster (CC) theory and present a pilot implementation for the Coupled Cluster Doubles method. We discuss the iterative solution of the CC amplitude equations using tensors in CP representation and present a tensor contraction scheme that minimizes the effort necessary for the rank reductions during the iterations. Furthermore, several details concerning the reduction of complexity of the algorithm, convergence of the CC iterations, truncation errors, and the choice of threshold for chemical accuracy are discussed.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Cluster, classify, regress: A general method for learning discontinuous functions

This paper presents a method for solving the supervised learning problem in which the output is highly nonlinear and discontinuous. Here, it is proposed to solve this problem in three stages: (i) cluster the pairs of input-output data points, resulting in a label for each point; (ii) classify the data, where the corresponding label is the output; and finally (iii) perform one separate regression for each class, where the training data corresponds to the subset of the original input-output pairs which have that label according to the classifier. It has not yet been proposed to combine these 3 fundamental building blocks of machine learning in this simple and powerful fashion. This can be viewed as a form of deep learning, where any of the intermediate layers can itself be deep. The utility and robustness of the methodology is illustrated on some toy problems, including one example problem arising from simulation of plasma fusion in a tokamak.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Vibrational frequencies for Be3 and Be4

The harmonic frequencies of Be3 and Be4, evaluated by multireference configuration-interaction (MRCI) and singles and doubles coupled-cluster (CCSD) methods, are reported, as are IR intensities evaluated by using the double harmonic approximation. A comparison of the results shows that the CCSD method augmented with a correction for connected triple excitations, CCSD(T), can be a viable alternative to MRCI for determining the ground-state properties of small Be clusters (other than the dimer). Results of normal mode analyses indicate that the anharmonic vibrational frequencies of Be3 and Be4 are reasonably consistent at the CCSD, CCSD(T), or MRCI levels of treatment.

Rendell, Alistair P.↗

Particle hit clustering and identification using point set transformers in liquid argon time projection chambers

Liquid argon time projection chambers are often used in neutrino physics and dark-matter searches because of their high spatial resolution. The images generated by these detectors are extremely sparse, as the energy values detected by most of the detector are equal to 0, meaning that despite their high resolution, most of the detector is unused in a particular interaction. Instead of representing all of the empty detections, the interaction is usually stored as a sparse matrix, a list of detection locations paired with their energy values. Traditional machine learning methods that have been applied to particle reconstruction such as convolutional neural networks (CNNs), however, cannot operate over data stored in this way and therefore must have the matrix fully instantiated as a dense matrix. Operating on dense matrices requires a lot of memory and computation time, in contrast to directly operating on the sparse matrix. We propose a machine learning model using a point set neural network that operates over a sparse matrix, greatly improving both processing speed and accuracy over methods that instantiate the dense matrix, as well as over other methods that operate over sparse matrices. Compared to competing state-of-the-art methods, our method improves classification performance by 14%, segmentation performance by more than 22%, while taking 80% less time and using 66% less memory. Compared to state-of-the-art CNN methods, our method improves classification performance by more than 86%, segmentation performance by more than 71%, while reducing runtime by 91% and reducing memory usage by 61%.

calibration and fitting methods↗

An Integrated Approach to Predictive Genomic Analytics

A variety of methods and algorithms have recently been employed in the analysis of gene expression data, including reverse-engineering and knowledge-based pathway modeling, semantic gene similarity, network analysis and clustering. These methods and algorithms address different subparts of the same overall challenge and need to be applied in combination to address predictive genomic analysis as a whole. In this paper, we present an integrated approach to predictive genomic analysis that achieves this objective and describe an application of the approach to the study of neuroprotection in stroke.

59 BASIC BIOLOGICAL SCIENCES↗

Sequencing the Black Aspergilli species complex

The ~15 members of the Aspergillus section Nigri species complex (the "Black Aspergilli") are significant as platforms for bioenergy and bioindustrial technology, as members of soil microbial communities and players in the global carbon cycle, and as food processing and spoilage agents and agricultural toxigens. Despite their utility and ubiquity, the morphological and metabolic distinctiveness of the complex's members, and thus their taxonomy, is poorly defined. We are using short read pyrosequencing technology (Roche/454 and Illumina/Solexa) to rapidly scale up genomic and transcriptomic analysis of this species complex. To date we predict 11197 genes in Aspergillus niger, 11624 genes in A. carbonarius, and 10845 genes in A. aculeatus. A. aculeatus is our most recent genome, and was assembled primarily from 454-sequenced reads and annotated with the aid of >2 million 454 ESTs and >300 million Solexa ESTs. To most effectively deploy these very large numbers of ESTs we developed 2 novel methods for clustering the ESTs into assemblies. We have also developed a pipeline to propose orthologies and paralogies among genes in the species complex. In the near future we will apply these methods to additional species of Black Aspergilli that are currently in our sequencing pipeline.

99 GENERAL AND MISCELLANEOUS↗

The peculiar velocities of rich clusters in the hot and cold dark matter scenarios

We present the results of a study of the peculiar velocities of rich clusters of galaxies. The peculiar motion of rich clusters in various cosmological scenarios is of interest for a number of reasons. Observationally, one can measure the peculiar motion of clusters to greater distances than galaxies because cluster peculiar motions can be determined to greater accuracy. One can also test the slope of distance indicator relations using clusters to see if galaxy properties vary with environment. We have used N-body simulations to measure the amplitude and rms cluster peculiar velocity as a function of bias parameter in the hot and cold dark matter scenarios. In addition to measuring the mean and rms peculiar velocity of clusters in the two models, we determined whether the peculiar velocity vector of a given cluster is well aligned with the gravity vector due to all the particles in the simulation and the gravity vector due to the particles present only in the clusters. We have investigated the peculiar velocities of rich clusters of galaxies in the cold dark matter and hot dark matter galaxy formation scenarios. We have derived peculiar velocities and associated errors for the scenarios using four values of the bias parameter ranging from b = 1 to b = 2.5. The growth of the mean peculiar velocity with scale factor has been determined and compared to that predicted by linear theory. In addition, we have compared the orientation of force and velocity in these simulations to see if a program such as that proposed by Bertschinger and Dekel (1989) for elliptical galaxy peculiar motions can be applied to clusters. The method they describe enables one to recover the density field from large scale redshift distance samples. The method makes it possible to do this when only radial velocities are known by assuming that the velocity field is curl free. Our analysis suggests that this program if applied to clusters is only realizable for models with a low value of the bias parameter, i.e., models in which the peculiar velocities of clusters are large enough that the errors do not render the analysis impracticable.

Rhee, George F.↗

Towards Effective Clustering Techniques for the Analysis of Electric Power Grids

Clustering is an important data analysis technique with numerous applications in the analysis of electric power grids. Standard clustering techniques are oblivious to the rich structural and dynamic information available for power grids. Therefore, by exploiting the inherent topological and electrical structure in the power grid data, we propose new methods for clustering with applications to model reduction, locational marginal pricing, phasor measurement unit (PMU or synchrophasor) placement, and power system protection. We focus our attention on model reduction for analysis based on time-series information from synchrophasor measurement devices, and spectral techniques for clustering. By comparing different clustering techniques on two instances of realistic power grids we show that the solutions are related and therefore one could leverage that relationship for a computational advantage. Thus, by contrasting different clustering techniques we make a case for exploiting structure inherent in the data with implications for several domains including power systems.

clustering, k-means, power grid, graphs↗

Spatio-Temporal Surrogates for Interaction of a Jet with High Explosives: Part II - Clustering Extremely High-Dimensional Grid-Based Data

Building an accurate surrogate model for the spatio-temporal outputs of a computer simulation is a challenging task. A simple approach to improve the accuracy of the surrogate is to cluster the outputs based on similarity and build a separate surrogate model for each cluster. This clustering is relatively straightforward when the output at each time step is of moderate size. However, when the spatial domain is represented by a large number of grid points, numbering in the millions, the clustering of the data becomes more challenging. In this report, we consider output data from simulations of a jet interacting with high explosives. These data are available on spatial domains of different sizes, at grid points that vary in their spatial coordinates, and in a format that distributes the output across multiple files at each time step of the simulation. We first describe how we bring these data into a consistent format prior to clustering. Borrowing the idea of random projections from data mining, we reduce the dimension of our data by a factor of thousand, making it possible to use the iterative k-means method for clustering. We show how we can use the randomness of both the random projections, and the choice of initial centroids in k-means clustering, to determine the number of clusters in our data set. Our approach makes clustering of extremely high dimensional data tractable, generating meaningful cluster assignments for our problem, despite the approximation introduced in the random projections.

97 MATHEMATICS AND COMPUTING↗

Clustering Analysis of Commercial Vehicles Using Automatically Extracted Features from Time Series Data

Standard of practice approaches to time series cluster analysis involve careful feature engineering, often utilizing expert input to tune and select features by hand. In many cases, expert input may not be readily available, or there may not yet exist a community consensus on the ideal features for a given application. This paper compares the results of several cluster analysis methods, using both hand selected features and those extracted automatically, when applied to large geospatial time series telematics data from commercial trucking fleets. The impacts of feature selection, dimensionality reduction, and choice of clustering algorithm on the quality of clustering results are explored. Results from this analysis confirm prior results that domain agnostic features are competitive with the hand engineered features with respect to clustering quality metrics. These results also provide new insight into the most successful strategies for identifying structure in large unstructured vehicle telematics data, and suggest that time series clustering using automatic feature extraction can be an effective approach to extract structure from large scale geospatial time series data in cases when hand selected features are not available.

33 ADVANCED PROPULSION SYSTEMS↗

Applications of Some Artificial Intelligence Methods to Satellite Soundings

Hard clustering of temperature profiles and regression temperature retrievals were used to refine the method using the probabilities of membership of each pattern vector in each of the clusters derived with discriminant analysis. In hard clustering the maximum probability is taken and the corresponding cluster as the correct cluster are considered discarding the rest of the probabilities. In fuzzy partitioned clustering these probabilities are kept and the final regression retrieval is a weighted regression retrieval of several clusters. This method was used in the clustering of brightness temperatures where the purpose was to predict tropopause height. A further refinement is the division of temperature profiles into three major regions for classification purposes. The results are summarized in the tables total r.m.s. errors are displayed. An approach based on fuzzy logic which is intimately related to artificial intelligence methods is recommended.

Munteanu, M. J.↗

Elaborated thermochemical treatment of HF, CO, N 2 , and H 2 O: Insight into HEAT and its extensions

Here, empirical, highly accurate non-relativistic electronic total atomization energies (eTAEs) are established by combining experimental or computationally converged treatments of the nuclear motion and relativistic contributions with the total atomization energies of HF, CO, N 2 , and H 2 O obtained from the Active Thermochemical Tables. These eTAEs, which have estimated (2σ) uncertainties of less than 10 cm -1 (0.12 kJ mol -1 ), form the basis for an analysis of high-level ab initio quantum chemical calculations that aim at reproducing these eTAEs for the title molecules. The results are then employed to analyze the performance of the high-accuracy extrapolated ab initio thermochemistry, or High-Accuracy Extrapolated Ab Initio Thermochemistry (HEAT), family of theoretical methods. The method known as HEAT-345(Q), in particular, is found to benefit from fortuitous error cancellation between its treatment of the zero-point energy, extrapolation errors in the Hartree-Fock and coupled cluster contributions, neglect of post-(T) core-correlation, and the basis-set error involved in higher-level correlation corrections. In addition to shedding light on a longstanding curiosity of the HEAT protocol—where the cheapest HEAT-345(Q) performs comparably to the theoretically more complete HEAT-456QP procedure—this study lays the foundation for extended HEAT variants that offer substantial improvements in accuracy relative to the established approaches.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Stochastic statistical theory of nucleation and evolution of nano-sized precipitates in alloys with application to precipitation of copper in iron

A consistent and computationally efficient stochastic statistical approach (SSA) is developed to study the kinetics of nucleation and evolution of nano-sized precipitates in alloys. To increase the accuracy of the method, many refinements of the previous simplified versions of this approach have been made. We consider a realistic vacancy-mediated exchange kinetics rather than the simplified direct-atomic-exchange model; use quantitative, cluster statistical methods rather than simple mean-field-type approximations; allow strong concentration and temperature dependences of generalized mobilities in the resulting kinetic equations; consider realistic alloy models based on first-principle calculations, and so on. We also introduce the 'maximum thermodynamic gain' principle to determine the key kinetic parameter of the SSA, the characteristic length of local equilibrium in the course of the nucleation process. For several realistic models of iron-copper alloys studied, the results of the SSA-based simulations of precipitation kinetics made in this work agree well with the kinetic Monte Carlo simulation results for all main characteristics of the microstructure. The approach developed is also used to study the kinetics of nucleation and changes in microstructural evolution under variations of temperature or concentration.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

The cluster-cluster correlation function

The clustering properties of the Abell and Zwicky cluster catalogs are studied using the two-point angular and spatial correlation functions. The catalogs are divided into eight subsamples to determine the dependence of the correlation function on distance, richness, and the method of cluster identification. It is found that the Corona Borealis supercluster contributes significant power to the spatial correlation function to the Abell cluster sample with distance class of four or less. The distance-limited catalog of 152 Abell clusters, which is not greatly affected by a single system, has a spatial correlation function consistent with the power law Xi(r) = 300r exp -1.8. In both the distance class four or less and distance-limited samples the signal in the spatial correlation function is a power law detectable out to 60/h Mpc. The amplitude of Xi(r) for clusters of richness class two is about three times that for richness class one clusters. The two-point spatial correlation function is sensitive to the use of estimated redshifts.

Postman, M.↗

Novel Histogram Based Unsupervised Classification Technique to Determine Natural Classes From Biophysically Relevant Fit Parameters to Hyperspectral Data

Hyperspectral image analysis has benefited from an array of methods that take advantage of the increased spectral depth compared to multispectral sensors; however, the focus of these developments has been on supervised classification methods. Lack of a priori knowledge regarding land cover characteristics can make unsupervised classification methods preferable under certain circumstances. An unsupervised classification technique is presented in this paper that utilizes physically relevant basis functions to model the reflectance spectra. These fit parameters used to generate the basis functions allow clustering based on spectral characteristics rather than spectral channels and provide both noise and data reduction. Histogram splitting of the fit parameters is then used as a means of producing an unsupervised classification. Unlike current unsupervised classification techniques that rely primarily on Euclidian distance measures to determine similarity, the unsupervised classification technique uses the natural splitting of the fit parameters associated with the basis functions creating clusters that are similar in terms of physical parameters. The data set used in this work utilizes the publicly available data collected at Indian Pines, Indiana. This data set provides reference data allowing for comparisons of the efficacy of different unsupervised data analysis. The unsupervised histogram splitting technique presented in this paper is shown to be better than the standard unsupervised ISODATA clustering technique with an overall accuracy of 34.3/19.0% before merging and 40.9/39.2% after merging. Finally, this improvement is also seen as an improvement of kappa before/after merging of 24.8/30.5 for the histogram splitting technique compared to 15.8/28.5 for ISODATA.

58 GEOSCIENCES↗

Automated Identification of Characteristic Droplet Size Distributions in Stratocumulus Clouds Utilizing a Data Clustering Algorithm

Abstract Droplet-level interactions in clouds are often parameterized by a modified gamma fitted to a “global” droplet size distribution. Do “local” droplet size distributions of relevance to microphysical processes look like these average distributions? This paper describes an algorithm to search and classify characteristic size distributions within a cloud. The approach combines hypothesis testing, specifically, the Kolmogorov–Smirnov (KS) test, and a widely used class of machine learning algorithms for identifying clusters of samples with similar properties: density-based spatial clustering of applications with noise (DBSCAN) is used as the specific example for illustration. The two-sample KS test does not presume any specific distribution, is parameter free, and avoids biases from binning. Importantly, the number of clusters is not an input parameter of the DBSCAN-type algorithms but is independently determined in an unsupervised fashion. As implemented, it works on an abstract space from the KS test results, and hence spatial correlation is not required for a cluster. The method is explored using data obtained from the Holographic Detector for Clouds (HOLODEC) deployed during the Aerosol and Cloud Experiments in the Eastern North Atlantic (ACE-ENA) field campaign. The algorithm identifies evidence of the existence of clusters of nearly identical local size distributions. It is found that cloud segments have as few as one and as many as seven characteristic size distributions. To validate the algorithm’s robustness, it is tested on a synthetic dataset and successfully identifies the predefined distributions at plausible noise levels. The algorithm is general and is expected to be useful in other applications, such as remote sensing of cloud and rain properties. Significance Statement A typical cloud can have billions of drops spread over tens or hundreds of kilometers in space. Keeping track of the sizes, positions, and interactions of all of these droplets is impractical, and, as such, information about the relative abundance of large and small drops is typically quantified with a “size distribution.” Droplets in a cloud interact locally, however, so this work is motivated by the question of whether the cloud droplet size distribution is different in different parts of a cloud. A new method, based on hypothesis testing and machine learning, determines how many different size distributions are contained in a given cloud. This is important because the size distribution describes processes such as cloud droplet growth and light transmission through clouds.

54 ENVIRONMENTAL SCIENCES↗

Cluster analysis of multiple planetary flow regimes

A modified cluster analysis method was developed to identify spatial patterns of planetary flow regimes, and to study transitions between them. This method was applied first to a simple deterministic model and second to Northern Hemisphere (NH) 500 mb data. The dynamical model is governed by the fully-nonlinear, equivalent-barotropic vorticity equation on the sphere. Clusters of point in the model's phase space are associated with either a few persistent or with many transient events. Two stationary clusters have patterns similar to unstable stationary model solutions, zonal, or blocked. Transient clusters of wave trains serve as way stations between the stationary ones. For the NH data, cluster analysis was performed in the subspace of the first seven empirical orthogonal functions (EOFs). Stationary clusters are found in the low-frequency band of more than 10 days, and transient clusters in the bandpass frequency window between 2.5 and 6 days. In the low-frequency band three pairs of clusters determine, respectively, EOFs 1, 2, and 3. They exhibit well-known regional features, such as blocking, the Pacific/North American (PNA) pattern and wave trains. Both model and low-pass data show strong bimodality. Clusters in the bandpass window show wave-train patterns in the two jet exit regions. They are related, as in the model, to transitions between stationary clusters.

Mo, Kingtse↗

Reward Driven Workflows for Unsupervised Explainable Analysis of Phases and Ferroic Variants From Atomically Resolved Imaging Data

Rapid progress in aberration corrected electron microscopy necessitates development of robust methods for the identification of phases, ferroic variants, and other pertinent aspects of materials structure from imaging data. While unsupervised methods for clustering and classification are widely used for these tasks, their performance can be sensitive to hyperparameter selection in the analysis workflow. In this study, the effects of descriptors and hyperparameters are explored on the capability of unsupervised ML methods to distill local structural information, exemplified by the discovery of polarization and lattice distortion in Sm − dopped BiFeO 3 (BFO) thin films. It is demonstrated that a reward-driven approach can be used to optimize these key hyperparameters across the full workflow, where rewards are designed to reflect domain wall continuity and straightness, ensuring that the analysis aligns with the material's physical behavior. This approach allows the discovery of local descriptors that are best aligned with the specific physical behavior, providing insight into the fundamental physics of materials. The reward driven workflow is further extended to disentangle structural factors of variation via an optimized variational autoencoder (VAE). Lastly, the importance of well-defined rewards is explored as a quantifiable measure of the success of the workflow.

Barakati, Kamyar [University of Tennessee, Knoxvil↗