Search

We present a new kernel-based algorithm for modeling evenly distributed multidimensional datasets that does not rely on input space sparsification. The presented method reorganizes the typical single-layer kernel-based model into a deep hierarchical structure, such that the weights of a kernel model over each dimension are modeled over its adjacent dimension. We show that modeling weights in the suggested structure leads to significant computational speedup and improved modeling accuracy.

The droplet model of the nucleus is revived, for which an exact solution for an incompressible fluid is obtained using the hydrodynamic potential solution obtained from the Schrödinger equation. Moreover, for an incompressible fluid, there are formulas for the pressure or potential. There is the main part of the hydrodynamic potential, which is obtained by replacing the modulus of the inverse difference of vectors by the difference in moduli of the values of the vectors. The bulk of the potential is expressed in a finite formula with singularities. A formula is obtained for the integral containing the modulus of the difference between the exact values of the vectors minus the main part of the potential. This difference defines a continuous correction with the features taken into account. The main part of the potential at the boundary of the nucleus turned out to be infinitely large with an imaginary part, locking particles in the nucleus. In this case, the real part of the main potential decreases with decreasing radius, becomes negative, and determines the bound state. At half the radius of the nucleus, there is a linear term along the radius. At the zero radius, there is an infinite negative potential with an imaginary part. An expression for the quantum of the emitted energy is obtained. Note that the added mass was not used due to the rotational regime of the nucleus. An algorithm for calculating the spectrum of the kernel is proposed, and each state of the action of the kernel sn corresponds to n calculated frequencies, determined by n angles in the configuration space. The main space is n + 1 dimensional, and each dimension of space has its own energy. But without special means, the potential of the nucleus tends to infinity. It is necessary to introduce the imaginary degree of roughness of the corners, in expressions containing singularities, then the infinities disappear.

According to Fisher's principles, an experimental field is typically divided into multiple blocks for local control. Although homogeneity is supposed within a block, this assumption may not be practical for large blocks, such as those including hundreds of plots. In line evaluation trials, which are essential in plant breeding, field heterogeneity must be carefully treated, because it can cause bias in the estimation of genetic potential. To more accurately estimate genotypic values in a large field trial, we developed spatial kernel models incorporating genome-wide markers, which consider continuous heterogeneity within a block and over the field. In the simulation study, the spatial kernel models were robust under various conditions. Although heritability, spatial autocorrelation range, replication number, and missing plots directly affected the estimation accuracy of genotypic values, the spatial kernel models always showed superior performance over the classical block model. We also employed these spatial kernel models for quantitative trait locus mapping. Finally, using field experimental data of bioenergy sorghum lines, we validated the performance of the spatial kernel models. The results suggested that a spatial kernel model is effective for evaluating the genetic potential of lines in a heterogeneous field.

Batched data is a type of data where each observed data value is the sum of a number of grouped (batched) latent ones obtained under different conditions. Batched data arises in various practical backgrounds and is often found in social studies and management sector. The analysis of such data is analytically challenging due to its structural complexity. In this article, we describe how to analyze batched service time data, estimate the mean and variance of each batch that are latent. We in particular focus on the situation when the observed total time includes an unknown proportion of non-service time. To address this problem, we propose a Gaussian model for efficiency as well as a semi-parametric kernel density model for robustness. We evaluate the performance of both proposed methods through simulation studies and then applied our methods to analyze a batched data.

In this paper, a displacement–pressure (u–p) semi-Lagrangian reproducing kernel (RK) is introduced to study the penetration depth of various projectile types into dry and fully saturated geo-materials. To describe the poromechanics of saturated materials, Biot theory is incorporated into the semi-Lagrangian formulation and a damage model is embedded into Drucker–Prager constitutive model to simulate the soil behavior and separation during the impact and penetration process. The stabilized nodal domain integration is developed in the two-field semi-Lagrangian RK formulation to ensure numerical stability, and the kernel contact algorithm is implemented to model the interaction between soil and projectile bodies. Several examples are studied to validate and assess the proposed method’s performance in predicting the final penetration depth, and the results are compared to those reported in the literature.

This work proposes a new approach for the mapping of a hysteresis curve. The idea developed is based on the use of PSO (Particle Swarm Optimization) to approximate the response of two well-known hysteresis models. In this sense, the PSO seeks to obtain a polynomial expression in the form of g(x,y) that performs the mapping of the L2P Kernel model, so that its response comes close to the response of a Presaich model with the same sign of excitement. The results provide a satisfactory preliminary response according to the similarities obtained from the dinamics of the two curves where they were very close, this shows the feasibility of using PSO algorithm to deform the curve L2P Kernel to obtain another existing hysteresis model, requiring to obtain a small number of parameters in the process.

This paper explores a succinct kernel model for Group-Sparse Projections Learning (GSPL), to handle multiview feature selection task completely. Compared to previous works, our model has the following useful properties: 1) Strictness: GSPL innovatively learns group-sparse projections strictly on multiview data via ‘2;0-norm constraint, which is different with previous works that encourage group-sparse projections softly. 2) Adaptivity: In GSPL model, when the total number of selected features is given, the numbers of selected features of different views can be determined adaptively, which avoids artificial settings. Besides, GSPL can capture the differences among multiple views adaptively, which handles the inconsistent problem among different views. 3) Succinctness: Except for the intrinsic parameters of projection-based feature selection task, GSPL does not bring extra parameters, which guarantees the applicability in practice. To solve the optimization problem involved in GSPL, a novel iterative algorithm is proposed with rigorously theoretical guarantees. Experimental results demonstrate the superb performance of GSPL on synthetic and real datasets.

Mass spectrometry (MS) is widely used for material analysis in various applications including forensic science. This work explores computational techniques and develops an application called "MSpec" using suitable algorithms for extracting informative parts of the MS dataset that aims towards pen ink classification. The system is intended as a tool that is capable of giving preliminary answers for such forensic analyses of documentary evidence involving different pen-ink types on writing. Support Vector Machine (SVM) was implemented and compared with other machine learning techniques via systematic performance assessments. They were trained and tested using MS data acquired from 10 blue-ink ballpoint pen samples, which were pre-processed using optimized steps. The results show that the tested models performed well in classifying the pen ink samples, with the SVM cubic kernel model giving the highest accuracy of 96.0%. Furthermore, dimensionality reduction of the dataset through peak detection helps improve the classification accuracy.

This paper focuses on the analysis and quantitative characterization of the effect of saturation on the viscoelastic properties of human root dentin. Uniaxial compression tests under creep conditions have been performed on root molar dentin with tubules fully saturated with a viscous physiological fluid, as well as samples with non-saturated tubules (dry dentin samples). Blair-Rabotnov (BR) fraction-exponential model is used to characterize the overall viscoelastic properties of dentin and correlate them to the level of saturation. Experimental data are compared with theoretical predictions that interrelate the viscoelastic properties of saturated and dry specimens. The results show that saturation increases the viscous creep strains of dentin, which indicates a reduced capacity for stress relief. The uniaxial compression test under creep conditions, in combination with the BR kernel model, allows us to analyze the creep-relaxation behavior of dentin. © 2020 Elsevier Ltd. TC, SS, and IS gratefully acknowledge financial support from National Institute of Health (USA), grant 2R25GM061222-18 . DZ, PP, and MYG gratefully acknowledge financial support from Russian Foundation for Basic Research (Russia), research project No. 18-38-20097 .

Simple Summary The African citrus psyllid Trioza erytreae is a natural vector of Candidatus liberibacter spp., the causal agent of the “citrus greening disease”. Citrus greening has not yet been detected in Europe; however, it represents a serious threat to citrus production. In this work, we parameterize a series of models to predict the expected spread of T. erytreae in the Iberian Peninsula since its introduction in 2014. Although T. erytreae was first detected in northwestern Spain, its detection and rapid spread around Porto (in Portugal) shortly afterward suggests a second entry point or transport of infested plant material from Spain. Among the developed models, the one that covered the known spread of T. erytreae best after 5 years was the kernel model with two simultaneous entry points. The invaded area predicted beyond the observed spread strongly suggests a physical and/or bioclimatic barrier preventing further spread of T. erytreae. Further development and refinement of models are crucial to accurately predicting the potential future spread of T. erytreae throughout the Iberian Peninsula. Accurate models will aid the development of successful management and regulatory programs. Abstract Assessing the potential of spread of an introduced crop pest in a new country is crucial to anticipating its effects on crop production and deriving phytosanitary management toward reducing potential negative effects. Citrus production represents a key agricultural activity throughout the Mediterranean basin. The African citrus psyllid Trioza erytreae (del Guercio, 1918) (Hemiptera: Triozidae) is a natural vector of Candidatus liberibacter spp., the causal agent of the harmful disease huanglongbing (HLB) or “citrus greening disease”. In continental Europe, T. erytreae was detected for the first time in northwestern Spain in 2014. Pest risk analysis (PRA) approaches, such as modeling, consider both time and space components to predict the potential distribution of pests in a given region. In this work, we aim to parameterize a model able to predict the expected spread of T. erytreae in the Iberian Peninsula using three types of PRA models. The kernel model with two hypothetical entry points accurately predicted the distribution of T. erytreae with respect to latitude. This model should be further refined and validated to support decision-makers in the adoption of timely and successful management and regulatory measures against the spread of T. erytreae to other citrus-producing areas in Europe.

A commercial near-infrared (NIR) instrument for bulk samples and a modified Single Kernel Near-Infrared (SKNIR) instrument equipped with a visible or NIR spectrometer were studied as a way to measure damage levels caused by Sunn pest (SP) in wheat. Sunn pest causes damage by feeding on wheat berries and injecting a salivary enzyme damaging the gluten. For measurement of SP damage in bulk wheat, NIR calibration models developed for mixtures containing 0–10% and 0–100% sound and SP damaged wheat resulted in R2 of 0.25 and 0.89 and SECVs of 2.75 and 10.9, respectively. The 0–100% model was considered a qualitative measure of damage, but predictions were poor over 0–10% which is a critical range for commercial applications. Discrimination between single kernels of Sound and SP-damaged was typically good, with classification accuracy averaging ∼75% for both visible and NIR although some were poor, which greatly affected the average. Average classification accuracy was ∼85% for spectral data that contained kernels from all samples. While the potential for using visible or NIR spectroscopy was shown, results highlighted the need to develop a more robust SP classification model to further evaluate the single kernel model.

In this study, we compared the prediction accuracy of the main genotypic effect model (MM) without G×E interactions, the multi-environment single variance G×E deviation model (MDs), and the multi-environment environment-specific variance G×E deviation model (MDe) where the random genetic effects of the lines are modeled with the markers (or pedigree). With the objective of further modeling the genetic residual of the lines, we incorporated the random intercepts of the lines (l) and generated another three models. Each of these 6 models were fitted with a linear kernel method (Genomic Best Linear Unbiased Predictor, GB) and a Gaussian Kernel (GK) method. We compared these 12 model-method combinations with another two multi-environment G×E interactions models with unstructured variance-covariances (MUC) using GB and GK kernels (4 model-method). Thus, we compared the genomic-enabled prediction accuracy of a total of 16 model-method combinations on two maize data sets with positive phenotypic correlations among environments, and on two wheat data sets with complex G×E that includes some negative and close to zero phenotypic correlations among environments. The two models (MDs and MDE with the random intercept of the lines and the GK method) were computationally efficient and gave high prediction accuracy in the two maize data sets. Regarding the more complex G×E wheat data sets, the prediction accuracy of the model-method combination with G×E, MDs and MDe, including the random intercepts of the lines with GK method had important savings in computing time as compared with the G×E interaction multi-environment models with unstructured variance-covariances but with lower genomic prediction accuracy.

Pest risk analyses (PRAs) are conducted to determine whether an organism is a pest and whether and how it should be regulated. Estimation of the potential area of establishment and pest spread are key factors of this analysis. Tools for modelling and mapping of these key factors have to be quick and easily applicable for a wide variety of organisms with limited data for parameterization. For this purpose, a dispersal kernel model based on a 2Dt-distribution had been developed in a European Union project (PRATIQUE). The aim of the present study was the evaluation of this spread model hitherto tested on insects, plants, fungi and nematodes in order to determine its applicability to bacterial pests. Therefore, the potential distribution and spread of kiwifruit bacterial canker Pseudomonas syringae pv. actinidiae in Europe was investigated based on climatic suitability and host plant availability. The results of the modelling were compared with the spread history of the pest in Europe. It is shown that this generic spread model can also be applied to a bacterial pest.

Developmental dyslexia (DD) and developmental coordination disorder (DCD) are distinct diagnostic disorders. However, they also frequently co-occur and may share a common etiology. It was proposed conceptually a neural network framework that explains differences and commonalities between DD and DCD through impairments of distinct or intertwined cortico-subcortical connectivity pathways. The present study addressed this issue by exploring intrinsic cortico-striatal and cortico-cerebellar functional connectivity in a large (n = 136) resting-state fMRI cohort study of 8–12-year-old children with typical development and with DD and/or DCD. We delineated a set of cortico-subcortical functional circuits believed to be associated with the brain’s main functions (visual, somatomotor, dorsal attention, ventral attention, limbic, frontoparietal control, and default-mode). Next, we assessed, using general linear and multiple kernel models, whether and which circuits distinguished between the groups. Findings revealed that somatomotor cortico-cerebellar and frontoparietal cortico-striatal circuits are affected in the presence of DCD, including abnormalities in cortico-cerebellar connections targeting motor-related regions and cortico-striatal connections mapping onto posterior parietal cortex. Thus, DCD but not DD may be considered as an impairment of cortico-subcortical functional circuits.

The feature information of the local graph structure and the nodes may be over-smoothing due to the large number of encodings, which causes the node characterization to converge to one or several values. In other words, nodes from different clusters become difficult to distinguish, as two different classes of nodes with closer topological distance are more likely to belong to the same class and vice versa. To alleviate this problem, an over-smoothing algorithm is proposed, and a method of reweighted mechanism is applied to make the trade-off of the information representation of nodes and neighborhoods more reasonable. By improving several propagation models, including Chebyshev polynomial kernel model and Laplace linear 1st Chebyshev kernel model, a new model named RWGCN based on different propagation kernels was proposed logically. The experiments show that satisfactory results are achieved on the semi-supervised classification task of graph type data.

© 2020 by the authors., Assessing the potential of spread of an introduced crop pest in a new country is crucial to anticipating its effects on crop production and deriving phytosanitary management toward reducing potential negative effects. Citrus production represents a key agricultural activity throughout the Mediterranean basin. The African citrus psyllid Trioza erytreae (del Guercio, 1918) (Hemiptera: Triozidae) is a natural vector of Candidatusliberibacter spp., the causal agent of the harmful disease huanglongbing (HLB) or “citrus greening disease”. In continental Europe, T. erytreae was detected for the first time in northwestern Spain in 2014. Pest risk analysis (PRA) approaches, such as modeling, consider both time and space components to predict the potential distribution of pests in a given region. In this work, we aim to parameterize a model able to predict the expected spread of T. erytreae in the Iberian Peninsula using three types of PRA models. The kernel model with two hypothetical entry points accurately predicted the distribution of T. erytreae with respect to latitude. This model should be further refined and validated to support decision-makers in the adoption of timely and successful management and regulatory measures against the spread of T. erytreae to other citrus-producing areas in Europe., The authors are grateful to the Foundation for Science and Technology (FCT, Portugal), for financial support through national funds FCT/MCTES to CIMO (UIDB/00690/2020) and to the project PRE-HLB—Preventing HLB epidemics for ensuring citrus survival in Europe (H2020-SFS-2018-2 Topic SFS-05-2018-2019-2020, proj. No. 817526).

How to select the suitable parameters and kernel model is a very important problem for Twin Support vector Machines (TWSVM). In order to solve this problem, one solving algorithm called binary PSO for optimizing the parameters of multi-core Twin Support Vector Machines (BPSO-MTWSVM) is proposed in this paper. Firstly, introducing multiple kernel functions, the twin support vector machines based on multi-core is constructed. This strategy is a good way to solve the kernel model selection. However, it has added three adjustable parameters. In order to solve the parameters selection problem which contain TWSVM parameters and multi-core model parameters, binary PSO (BPSO) is introduced. BPSO is an optimization algorithm who has strong robustness and good global searching ability. Finally, compared with the classical TWSVM the experimental results show that BPSO-MTWSVM has higher classification accuracy.

Exploring the phases of matter is important in condensed matter physics and materials discovery, and here Raman spectroscopy gives a material's structural fingerprint. The authors develop a highly correlated kernel model based on a set of Raman-scattering spectra, to deduce thermally induced transitions among three phases in a model ferroelectric crystal, by mining and learning the phonon behaviors in the crystal lattice. The underlying physical mechanism behind these behaviors, and just how the model ``learns'' these rules to make precise predictions, are clarified. This insight paves the way to applying the generic approach for prediction of unexplored structures and materials.

There are four widely-used kernel driven models in the thermal infrared domain, including LSF-Li, RL, Ross-Li and Vinnikov. They have great potentials to achieve the angular effect correction for land surface temperature products. RL and Ross-Li are direct extensions of kernel models in the visible and near infrared region and other two models were directly designed for the thermal infrared region. Their fitting abilities should be evaluated comprehensively at first. Some evaluation works have been reported based on simulated datasets or measured dataset with limited view angles. In this paper, two measured multi-angle datasets with over 104 directions were selected to do evaluation. Both of them were extracted from the airborne measurement using spatial-temporal average method. We found that LSF-Li is always the best one, followed by Ross-Li, RL and Vinnikov.

Fumonisin mycotoxins are a persistent challenge to human and livestock health in tropical and sub-tropical maize cropping systems, and more efficient methods are needed to reduce their presence in food systems. We constructed a novel, low-cost device for sorting grain, the “DropSort”, and tested its effectiveness on both plastic kernel models and fumonisin-contaminated maize. Sorting plastic kernels of known size and shape enabled us to optimize the sorting performance of the DropSort. The device sorted maize into three distinct fractions as measured by bulk density and 100-kernel weight. The level of fumonisin was lower in the heaviest fractions of maize compared to the unsorted samples. Based on correlations among fumonisin and bulk characteristics of each fraction, we found that light fraction 100-kernel weight could be an inexpensive proxy for unsorted fumonisin concentration. Single kernel analysis revealed significant relationships among kernel fumonisin content and physical characteristics that could prove useful for future sorting efforts. The availability of a low-cost device (materials~USD 300) that can be used to reduce fumonisin in maize could improve food safety in resource-limited contexts in which fumonisin contamination remains a pressing challenge.

Endogenous bursters in central pattern generators (CPGs) generate rhythmic firing patterns controlling regular movements in the organism. Based on a pacemaker kernel model of the stomatogastric ganglion (SGG) of crustaceans, we constructed three reduced models, (i) dendrite-reduced model (DRM), (ii) axon-reduced model (ARM), and (iii) primary neuritereduced model (PNRM). Similar firing patterns were observed in two models except the axonreduced one. Perturbing of various parameters in the models induced bifurcation phenomena in the occurrence of interspike intervals (ISIs), which depicted variation of the firing patterns. By comparing and analyzing two-dimensional parameter planes derived from the above different models, the effects of compartments on varying firing patterns were detected. In particular, a different kind of period-doubling transition mode of firing patterns, which varied via a ring-shape mode, was found. Ендогенні компоненти центральних генераторів патернів (ЦГП), відповідальні за контроль певних стандартних моторних феноменів в організмі, генерують ритмічні групи розрядів. Базуючись на ядерній моделі пейсмекерів у стоматогастричному ганглії (СГГ) ракоподібних, ми сконструювали три редуковані моделі – модель з редукованими дендритами (ДРМ), модель з редукованим аксоном (АРМ) та модель з редукованим первинним нейритом (ПНРМ). У двох моделях, за виключенням АРМ, спостерігались однакові патерни розрядів. Змини різних параметрів моделей призводили до появи біфуркаційних феноменів у послідовностях міжімпульсних інтервалів, що віддзеркалювалось у варіаціях патернів розрядів. У перебігу порівняння двовимірних площин параметрів, отриманих для різних моделей, вдалось ідентифікувати впливи компартментів на варіацію параметрів розрядів. Зокрема, було встановлено специфічний вид перехідного режиму подвоєння періоду в патернах, варіації якого мали кільцеподібний характер.

Hydrogels are used in multiple industries; although hydrogel deformation and wear behaviors can be a good indicator for their functionality, there are no available models for characterizing hydrogel deformation–wear behaviors. This study characterized wear behaviors of casein (3–5% w/w, 200 mM NaCl), κ-carrageenan (0.8–1.2% w/w, 100 mM NaCl), and whey protein gels (13% w/w, 50–150 mM NaCl) at room temperature (20 °C) under several normal forces (0.1–1.5 N). A kernel-based model was used to characterize hydrogel wear behavior by separating the deformation–wear process deformation-dominant, constant wear rate, and failure regions. Higher normal force and lower polymer concentration showed higher constant wear rates and lower durability. The kernel model was able to characterize hydrogel wear-deformation behaviors and is useful for industrial applications.

We present several techniques for modeling emergency events using data from 911 emergency calls in the city of Cuenca-Ecuador. We apply three types of models. First, we use a probabilistic description of events using Gaussian kernels based on both, regular segmentation and mixture models, to represent the spatial distribution of occurrences. Second, we verify the qualitative relation of the clusters obtained with our kernel model with respect to the geo-political organization of the city. Finally, we develop an emergency model using a large dataset corresponding to the period January 1st 2015 through December 31st 2016 and test various data mining algorithms for prediction purposes. We verify the usefulness of our approach experimentally.

The advantage of using multifeature information for classification has been widely recognized. Representation-based methods with multifeature combination learning have only recently attracted increasing attention for hyperspectral classification. However, nonlinearity in data and the computational load of processing multifeature information and contextual information have been two thorny issues. In this paper, we present a fast joint sparse representation model with multifeature combination learning and its kernel extensions for hyperspectral imagery classification. For several complementary features (spectral, shape, and texture), the proposed model simultaneously acquires a representation vector for each type of feature and encourages the representation vectors to share a common sparsity pattern by imposing the joint sparsity ? r o w , 0 -norm regularization. Thus, the cross-feature information can be taken into account. For different features, different weights are assigned since they may not contribute equally to the final decision. Furthermore, kernel joint sparse representation model is presented to handle nonlinearity in the data. Kernel model projects the data into a high-dimensional space to improve the separability, achieving a better performance than the linear version. At the same time, we incorporate contextual neighborhood knowledge into the learned models. Experiments on several real hyperspectral images indicate that the proposed algorithms with much less memory requirements perform significantly faster than state-of-the-art algorithms, while exhibit highly competitive classification accuracy.

In this research, we study data modeling of Indonesian inflation in the transportation, communication and financial services sector using the kernel and spline models. Determination of the optimal models based on the smallest of GCV value and determination of the best model based on the smallest out sampels of Mean Square Error (MSE) value. By modeling the yoy (year on year) inflation data in Indonesia in the transportation, communication and financial services sector In January 2007 to January 2015, shows that the kernel model using Gaussian kernel function obtained optimal model with a bandwidth 0.24 and the optimal spline model with order 5 and 4 points knots. Based on out sampels data in February to August 2015, obtained out sampels MSE value of the spline model is smaller than the kernel model. So that the spline model is better than the kernel model to analyze the inflation data of transportation, communication and financial services sector. Keywords: Inflation, Transportation, Communication and Financial Services Sector, Kernel, Spline, GCV, MSE.

Aiming at the contradictory problem that the hotels’ daily excess ingredient causes a lot of losses, which is contradictory to the profit purpose of ingredient, this paper establishes kernel model and Extensible model through Extensible theory methods combined with mind mapping. Through evaluating the characteristics of alike relative matter-element, making up comprehensive dependent functions and using Extensible transformation, we can get various solutions. And then, taking the matter-element characteristics as evaluation criteria, we can carry out the superiority evaluation to the solutions and finally get the superior solution.