High Accuracy Of Method Research Articles

Gaussian processes for finite size extrapolation of many-body simulations.

Key to being able to accurately model the properties of realistic materials is being able to predict their properties in the thermodynamic limit. Nevertheless, because most many-body electronic structure methods scale as a high-order polynomial, or even exponentially, with system size, directly simulating large systems in their thermodynamic limit rapidly becomes computationally intractable. As a result, researchers typically estimate the properties of large systems that approach the thermodynamic limit by extrapolating the properties of smaller, computationally-accessible systems based on relatively simple scaling expressions. In this work, we employ Gaussian processes to more accurately and efficiently extrapolate many-body simulations to their thermodynamic limit. We train our Gaussian processes on Smooth Overlap of Atomic Positions (SOAP) descriptors to extrapolate the energies of one-dimensional hydrogen chains obtained using two high-accuracy many-body methods: coupled cluster theory and Auxiliary Field Quantum Monte Carlo (AFQMC). In so doing, we show that Gaussian processes trained on relatively short 10-30-atom chains can predict the energies of both homogeneous and inhomogeneous hydrogen chains in their thermodynamic limit with sub-milliHartree accuracy. Unlike standard scaling expressions, our GPR-based approach is highly generalizable given representative training data and is not dependent on systems' geometries or dimensionality. This work highlights the potential for machine learning to correct for the finite size effects that routinely complicate the interpretation of finite size many-body simulations.

High-accuracy quantification of soil elements by laser-induced breakdown spectroscopy based on PCA-GS-ELM

A new strategy to mitigate the impact of matrix effects and a high-accuracy quantification method for elements in soil by LIBS called PCA-GS-ELM are proposed.

FESD-J: Finite Elements with Switch Detection for numerical optimal control of rigid bodies with impacts and Coulomb friction

The Finite Elements with Switch Detection (FESD) is a high-accuracy method for the numerical simulation and solution of optimal control problems subject to discontinuous ODEs. In this article, we extend the FESD method (Nurkanović et al., 2022) to the dynamic equations of multiple rigid bodies that exhibit state jumps due to impacts and Coulomb friction. This new method is referred to as FESD with Jumps (FESD-J). Starting from the standard Runge–Kutta equations, we let the integration step sizes be degrees of freedom. Additional constraints are introduced to ensure exact switch detection and to remove spurious degrees of freedom if no switches occur. Moreover, at the boundaries of each integration interval (finite element), we impose the impact equations in their complementarity form, at both the position and velocity level. They compute the normal and tangential impulses in case of contact making. Otherwise, they are reduced to the continuity conditions for the velocities. FESD-J treats multiple contacts, where each contact can have a different coefficient of restitution and friction. All methods introduced in this paper are implemented in the open-source software package NOSNOC (Nurkanović and Diehl, 2022). We illustrate the use of FESD-J in both simulation and optimal control examples.

High-Accuracy Mapping of Soil Parent Material Types in Hilly Areas at the County Scale Using Machine Learning Algorithms

Conventional maps of soil parent material (SPM) types obtained by field survey and manual mapping or predictions from other map data have limited accuracy. Digital soil mapping of SPM types necessitates accurate acquisition of SPM distribution information, which is still a challenge in hilly areas. This study developed a high-accuracy method for SPM identification in hilly areas at the county scale. Based on geographic information system technology, seven feature variables were extracted from the geological map, geomorphic map, digital elevation model, and remote sensing image data of Shanggao County, Jiangxi Province, China. Different feature combination schemes were designed to develop SPM identification models based on random forest (RF), support vector machine (SVM), and maximum likelihood classification (MLC) algorithms. The best SPM identification results were obtained from the RF algorithm using the combination of geological type, geomorphic type, elevation, and slope. Confusion matrices were constructed based on a field survey of 586 validation samples, and the results were evaluated in terms of overall accuracy, precision, recall, F1 score, and Kappa coefficient. The overall accuracy and Kappa coefficient of the results from the optimal RF model were 83.11% and 0.79, respectively, which were 26.11% and 0.31 higher than those of the conventional map, respectively. Its precision and recall for various SPM types were greater than 75%. A comprehensive comparison of the accuracy, uncertainty, and plotting performance of the SPM recognition results reveals that the RF algorithm outperforms the SVM algorithm and the MLC algorithm. Geological type was the largest contributor to SPM identification, followed by geomorphic type, elevation, and slope. The importance of different feature variables varied for distinct SPM types. The accuracy of SPM identification was not improved by selecting more feature variables, such as land use type, normalised difference vegetation index, and topographic wetness index. This study demonstrates the feasibility of high-accuracy county-level SPM mapping in hilly areas based on the RF algorithm using geological type, geomorphic type, elevation, and slope as feature variables. As hilly areas have typical topographic features and SPM types, the proposed method of SPM mapping can be useful for application in other similar areas. There are a few limitations in this study with regard to data quality and resolution, feature variable selection, classification algorithm generalisation, and study area representativeness, which may affect the outcomes and need to be solved.

A Computational Conformal Geometry Approach to Calculate the Large Deformation of Plates/Shells With Arbitrary Shapes

Abstract High-accuracy numerical methods to solve the nonlinear Föppl–von Kármán (FvK) equations usually work well only in simple domains such as rectangular regions. Computational conformal geometry (CCG) provides a systematic method to transform complicated surfaces into simple domains, preserving the orthogonal frames such that the corresponding FvK equations can be solved by more effective numerical methods. Based on CCG, we proposed a general method for solving large deformation and nonlinear vibration of plate/shell structures with arbitrary shapes. The method can map any complex surface conformal to a rectangular region, and then FvK equations are solved in the rectangular region to study nonlinear vibration problems of any arbitrary shape plates/shells. The conform map is calculated by solving Laplace equations on a fine Delauney triangular mesh on the surface, which is numerically robust, and the map is harmonic and subsequently C∞ smooth, such that all the evaluations and spatial derivatives required by high accuracy methods at the regular nodes can be accurately and efficiently calculated. A variational function that is equivalent to the FvK equations is provided, such that the FvK equations can be solved by multiple numerical methods. The degree-of-freedom in solving the FvK equations is usually much less than that in the finite element methods described by displacements. The effectiveness of the proposed approach is verified by several benchmark examples, and the current method is suitable for calculating the large deflections and nonlinear dynamical responses of plates/shallow shells with arbitrary shapes.

How to Explain Sentiment Polarity – A Survey of Explainable Sentiment Analysis Approaches

Sentiment analysis area has become more and more popular due to information and opinion overload (especially in social networks). With growth of efficient and high accuracy methods of artificial intelligence, it is easier to obtain satisfactory results using opinion mining. Usually, more complex structure, and more sophisticated methods give us better results, but it is hard to understand why such a result was returned. Many of these methods are black-box models – it is a reason for decreasing trust in the system. To manage the problem, explainable methods were developed. This paper presents a systematic survey of explainable artificial intelligence methods that are used in sentiment analysis area. We have analyzed and classified existing approaches to Explainable Sentiment Analysis (XSA) and indicated trends and challenges in this area.

NEURAL NETWORK METHOD FOR PARAMETER ESTIMATION OF FRACTIONAL DISCRETE-TIME UNIFIED SYSTEMS

Data-driven learning of the fractional discrete-time unified system is studied in this paper. A neural network method is suggested in the parameter estimation of fractional discrete-time chaotic systems. An optimization problem is obtained and the famous Adam algorithm is employed to train the neural network’s weights and parameters. The parameter estimation result is compared with that of the stepwise response sensitivity approach (SRSA). This paper provides a high accuracy method for parameter inverse problems. The method also can be applied to data-driven learning of other fractional chaotic systems.

Optimal self-stress determination for high-accuracy mesh reflectors design considering the pillow distortion

The pillow distortion is a significant barrier to the high-accuracy design of mesh reflector antennas, which is caused by the compatible deformation between the flexible cable network and the metallic mesh, manifested as the pillow-like distortion of the mesh reflecting surface. The pillow effect deviates the mesh surface from the desired paraboloidal shape leading to the surface error. Unfortunately, current researches usually neglect this effect in the mesh reflector design due to its complicated formulation. Focusing on this phenomenon, this paper proposed an optimal self-stress determination method for high-accuracy mesh reflector design. Firstly, a mechanism model of pillow distortion is proposed based on the cable-membrane constitutive models and coupling relations. The integrated equilibrium equation of the mesh reflector is established and solved by the dynamic relaxation (DR) method, which effectively analyses the influence of the pillow distortion on the mesh reflector. Then, considering the pillow distortion, a high-accuracy design method for the mesh reflector is proposed by optimising the geometrical sizes of the cable network and the metallic mesh combining the DR method with optimisation algorithm. Finally, the proposed method is effectively applied to the design of an eighteen-unit mesh reflector antenna, and the feasibility and effectiveness of the proposed method are demonstrated.

Simulation and experimental research on the abrasive trajectories of plane lapping based on rotary swing drive

In the pursuit of achieving exceptional surface shape accuracy for MPCVD (Microwave Plasma Chemical Vapor Deposition) polycrystalline diamond wafers during the free abrasive lapping process, plane lapping based on rotary swing drive was employed in this study. The crux of this study lay in the development of a comprehensive kinematic model, scrutinizing the motion behavior of abrasive particles, and probing the influence of parameters on the uniformity of abrasive trajectories. The results show that the eccentricity, arc chord length and speed ratio affect the probability of track coincidence rate, and then affect the distribution uniformity. And then the verification experiments solidifying the reliability of our kinematic model and affirming the veracity of simulation results. The best surface shape accuracy ( PV) of the MPCVD polycrystalline diamond wafer was 2.3 μm under optimal parameters, representing a substantial advancement compared to the original lapping method, which only yielded a PV value of 8.4 μm. This study provides a promising method for high surface shape accuracy of MPCVD polycrystalline diamond wafers with large sizes.

Aberrant landscapes of maternal meiotic crossovers contribute to aneuploidies in human embryos.

Meiotic recombination is crucial for human genetic diversity and chromosome segregation accuracy. Understanding its variation across individuals and the processes by which it goes awry are long-standing goals in human genetics. Current approaches for inferring recombination landscapes rely either on population genetic patterns of linkage disequilibrium (LD)-capturing a time-averaged view-or on direct detection of crossovers in gametes or multigeneration pedigrees, which limits data set scale and availability. Here, we introduce an approach for inferring sex-specific recombination landscapes using data from preimplantation genetic testing for aneuploidy (PGT-A). This method relies on low-coverage (<0.05×) whole-genome sequencing of in vitro fertilized (IVF) embryo biopsies. To overcome the data sparsity, our method exploits its inherent relatedness structure, knowledge of haplotypes from external population reference panels, and the frequent occurrence of monosomies in embryos, whereby the remaining chromosome is phased by default. Extensive simulations show our method's high accuracy, even at coverages as low as 0.02×. Applying this method to PGT-A data from 18,967 embryos, we mapped 70,660 recombination events with ∼150 kbp resolution, replicating established sex-specific recombination patterns. We observed a reduced total length of the female genetic map in trisomies compared with disomies, as well as chromosome-specific alterations in crossover distributions. Based on haplotype configurations in pericentromeric regions, our data indicate chromosome-specific propensities for different mechanisms of meiotic error. Our results provide a comprehensive view of the role of aberrant meiotic recombination in the origins of human aneuploidies and offer a versatile tool for mapping crossovers in low-coverage sequencing data from multiple siblings.

A SPH method of high accuracy and efficiency for low and medium Reynolds number flow problems

A SPH method of high accuracy and efficiency for low and medium Reynolds number flow problems

DecOM: similarity-based microbial source tracking of ancient oral samples using k-mer-based methods

BackgroundThe analysis of ancient oral metagenomes from archaeological human and animal samples is largely confounded by contaminant DNA sequences from modern and environmental sources. Existing methods for Microbial Source Tracking (MST) estimate the proportions of environmental sources, but do not perform well on ancient metagenomes. We developed a novel method called decOM for Microbial Source Tracking and classification of ancient and modern metagenomic samples using k-mer matrices.ResultsWe analysed a collection of 360 ancient oral, modern oral, sediment/soil and skin metagenomes, using stratified five-fold cross-validation. decOM estimates the contributions of these source environments in ancient oral metagenomic samples with high accuracy, outperforming two state-of-the-art methods for source tracking, FEAST and mSourceTracker.ConclusionsdecOM is a high-accuracy microbial source tracking method, suitable for ancient oral metagenomic data sets. The decOM method is generic and could also be adapted for MST of other ancient and modern types of metagenomes. We anticipate that decOM will be a valuable tool for MST of ancient metagenomic studies.6i9WtiAouxwxZ9D7DN1ayhVideo

An Encoder-Based Relative Attitude Observation Method for Self-Calibration in Dual-Axis RINS

This study investigates an encoder-based relative attitude observation method for self-calibration in dual-axis rotational inertial navigation system (RINS). First, the calibration error of the gyroscope parameters caused by the velocity and position observations in systematic calibration is analyzed. Second, a relative attitude observation method utilizing photoelectric encoders for gyroscope parameters estimation is proposed. We apply the proposed method in our designed dual-axis RINS to verify its effectiveness in this work. Experimental results show that the calibration accuracy of the gyroscope can be improved to 0.0005 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$^ \circ /h$</tex-math></inline-formula> in bias, and 0.98 <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"><tex-math notation="LaTeX">$ppm$</tex-math></inline-formula> in scale factor using the proposed method. The proposed self-calibration method can be utilized as a high-accuracy self-calibration method in dual-axis RINS when the vehicle is in static condition.

Fast autoregressive tensor decomposition for online real-time traffic flow prediction

Online real-time traffic flow prediction typically offers better real-time performance than offline prediction. However, existing studies rarely discussed online real-time traffic flow prediction and the balance between prediction accuracy and computational costs. Training and predicting traffic flow data with complex patterns by artificial intelligence models is usually time-consuming. Some high-accuracy statistical learning methods also reach a performance bottleneck in terms of computational speed. Therefore, a Fast Autoregressive Tensor Decomposition (FATD) algorithm is proposed for online real-time traffic flow prediction. First, the historical tensor data is decomposed by Tucker Decomposition into factor matrices and easy-to-compute core tensors, and these core tensors are modeled by Tensor Seasonal Autoregressive Integrated Moving Average (Tensor SARIMA). Second, a future core tensor is predicted by Tensor SARIMA. By the Inverse Tucker Decomposition, the traffic flow data to be predicted is recovered. The experimental results show that the FATD algorithm can reduce the computational cost while maintaining high accuracy. Compared with baselines, the FATD algorithm reduces MAE, RMSE, and the computational cost by approximately 35.04%, 26.86%, and 99.28% on average, respectively.

High-accuracy vanishing-constraints-based calibration of fringe projection vision sensor

In fringe projection profilometry (FPP), calibration of the vision sensor is crucial to ensure the accuracy of three-dimensional (3D) shape measurement. However, due to correspondence errors of planar calibration target (PCT) 3D points and its image points, and low mapping accuracy between projector pixels and camera pixels, traditional calibration methods based on the PCT sometimes cannot provide reliable and accurate results. Aiming at these problems, a high-accuracy calibration method based on vanishing constraints for fringe projection vision sensor (FPVS) is proposed in this paper. The intrinsic parameters are calibrated using the constraints of vanishing points, the rotation matrix is calibrated using the constraints of vanishing lines, and the optimal calibration parameters are estimated by the objective function established based on the 2D and 3D constraints of feature points. Experimental results demonstrated the performance of our calibration method.

High-accuracy optical coherence elastography digital volume correlation methods to measure depth regions with low correlation.

The decreasing correlation of optical coherence tomography (OCT) images with depth is an unavoidable problem for the depth measurement of the digital volume correlation (DVC) based optical coherence elastography (OCE) method. We propose an OCE-DVC method to characterize biological tissue deformation in deeper regions. The method proposes a strategy based on reliability layer guided displacement tracking to achieve the OCE-DVC method for the deformation measurement in deep regions of OCT images. Parallel computing solves the computational burden associated with the OCE-DVC method. The layer-by-layer adaptive data reading methods are used to guarantee the parallel computing of high-resolution OCT images. The proposed method shown in this study nearly doubles the depth of quantitative characterization of displacement and strain. At this depth, the standard deviation of displacement and strain measurements is reduced by nearly 78%. Under nonuniform deformation field, OCE-DVC method tracked the displacement with large strain gradient in depth region.

Sim-GAIL: A generative adversarial imitation learning approach of student modelling for intelligent tutoring systems

The continuous application of artificial intelligence (AI) technologies in online education has led to significant progress, especially in the field of Intelligent Tutoring Systems (ITS), online courses and learning management systems (LMS). An important research direction of the field is to provide students with customised learning trajectories via student modelling. Previous studies have shown that customisation of learning trajectories could effectively improve students’ learning experiences and outcomes. However, training an ITS that can customise students’ learning trajectories suffers from cold-start, time-consumption, human labour-intensity, and cost problems. One feasible approach is to simulate real students’ behaviour trajectories through algorithms, to generate data that could be used to train the ITS. Nonetheless, implementing high-accuracy student modelling methods that effectively address these issues remains an ongoing challenge. Traditional simulation methods, in particular, encounter difficulties in ensuring the quality and diversity of the generated data, thereby limiting their capacity to provide intelligent tutoring systems (ITS) with high-fidelity and diverse training data. We thus propose Sim-GAIL, a novel student modelling method based on generative adversarial imitation learning (GAIL). To the best of our knowledge, it is the first method using GAIL to address the challenge of lacking training data, resulting from the issues mentioned above. We analyse and compare the performance of Sim-GAIL with two traditional Reinforcement Learning-based and Imitation Learning-based methods using action distribution evaluation, cumulative reward evaluation, and offline-policy evaluation. The experiments demonstrate that our method outperforms traditional ones on most metrics. Moreover, we apply our method to a domain plagued by the cold-start problem, knowledge tracing (KT), and the results show that our novel method could effectively improve the KT model’s prediction accuracy in a cold-start scenario.

Classification of COVID-19 and lung opacity using vision transformer on chest x-ray images

There are several recent works which had proposed an automatic computer-aided diagnosis (CAD) deep learning (DL) model to diagnose coronavirus disease 2019 (COVID-19) using chest x-ray images (CXR) to propose a high-accuracy CAD method to detect COVID-19 automatically. In this study, seven different models including Convolutional Neural Network (CNN) models such as VGG-16 and vision transformer (ViT) models, are proposed. The different proposed models are trained with a three-class balanced dataset consisting of 3,000 CXR images consisting of 1,000 CXR images for each class of COVID-19, Normal, and Lung-Opacity. A publicly available dataset to train and test the models is used from Kaggle-COVID-19-Radiography-Dataset. From the experiments, the accuracy of the VGG16 model is 93.44% and ViT’s is 92.33%. Besides, the binary classification between two classes of COVID-19 and Normal CXR with a limited number of just 100 images for each class, using a transfer learning technique, with a validation accuracy of 97.5% is proposed.

Testability design of power distribution system considering noise immunity characteristics

MOSFET-based power electronics systems have high-reliability requirements, and a high-accuracy fault diagnosis method is necessary. However, imperfect conditions such as severe signal noise and poor consistency exist in actual products. Conventional fault diagnosis methods are difficult to identify faults accurately. This paper describes a simulation-based testability design method for batch products considering noise immunity. First, the components' soft and hard fault models are determined with failure mode and effect analysis (FMEA), and the circuit simulation model is constructed. Then, the fault injection simulation is performed to structure the dependency matrix. The test points scheme is selected and optimized by extracting noise immunity characteristics of different types of signals. Finally, a power distribution system is taken as a case study for the testability design method, and the effectiveness of the test points is evaluated through simulation and experiment verification.

High accurate detection method for aortic valve opening of seismocardiography signals

Accurately detecting the aortic valve opening (AO) peaks of seismocardiography (SCG) is a challenging problem due to interference of other morphological inflection points. In this paper, a high accurate method is proposed to extract AO peaks based solely upon the SCG data. The raw SCG signal is purified by a simple first order interference cancellation method, resulting in a reduced number of modes to be decomposed. The purified SCG signal is then decomposed into a series of quasi-orthogonal modes by successive variational mode decomposition (SVMD) without any prior about the number of modes. Considering the pulsatile nature of the AO signal, a waveform factor criterion is proposed to reconstruct the AO signal based on the pulsatile level of each mode. A seventh power law detector is designed to amplify the AO peak and suppress spurious peaks. The publicly available combined measurement of electrocardiogram (ECG), breathing and seismocardiograms (CEBS) database is exploited to verify the performance of the proposed method. We show that the average sensitivity of our technique is 99.02%, the prediction rate is 99.06%, and the detection accuracy is 98.10%, which is superior to several state-of-the-art methods. In addition, compared with the ECG reference value, the instantaneous heart rate extracted by the proposed method is in good agreement with that of the ECG, e.g., the maximum average absolute error percentage is as low as 2.11 and the maximum average value relative error is about 0.03, further demonstrating that the proposed method can achieve accurate estimates of heart rate with an accelerometer alone.