P. K. A. Chitra¹, S. Appavu alias Balamurugan², S. Geetha³, Seifedine Kadry^4,5,6, Jungeun Kim^7,*, Keejun Han⁸

Computer Systems Science and Engineering, Vol.48, No.5, pp. 1367-1385, 2024, DOI:10.32604/csse.2023.034373 - 13 September 2024

Abstract A generalization of supervised single-label learning based on the assumption that each sample in a dataset may belong to more than one class simultaneously is called multi-label learning. The main objective of this work is to create a novel framework for learning and classifying imbalanced multi-label data. This work proposes a framework of two phases. The imbalanced distribution of the multi-label dataset is addressed through the proposed Borderline MLSMOTE resampling method in phase 1. Later, an adaptive weighted l₂₁ norm regularized (Elastic-net) multi-label logistic regression is used to predict unseen samples in phase 2. The proposed… More >

Zi Han^1,*, Zhentian Huang²

CMES-Computer Modeling in Engineering & Sciences, Vol.138, No.1, pp. 245-272, 2024, DOI:10.32604/cmes.2023.029708 - 22 September 2023

Abstract In the context of global mean square error concerning the number of random variables in the representation, the Karhunen–Loève (KL) expansion is the optimal series expansion method for random field discretization. The computational efficiency and accuracy of the KL expansion are contingent upon the accurate resolution of the Fredholm integral eigenvalue problem (IEVP). The paper proposes an interpolation method based on different interpolation basis functions such as moving least squares (MLS), least squares (LS), and finite element method (FEM) to solve the IEVP. Compared with the Galerkin method based on finite element or Legendre polynomials,… More > Graphic Abstract

Supriya Gupta^*, Aakanksha Sharaff, Naresh Kumar Nagwani

Computer Systems Science and Engineering, Vol.45, No.3, pp. 2333-2349, 2023, DOI:10.32604/csse.2023.030385 - 21 December 2022

Abstract Text Summarization models facilitate biomedical clinicians and researchers in acquiring informative data from enormous domain-specific literature within less time and effort. Evaluating and selecting the most informative sentences from biomedical articles is always challenging. This study aims to develop a dual-mode biomedical text summarization model to achieve enhanced coverage and information. The research also includes checking the fitment of appropriate graph ranking techniques for improved performance of the summarization model. The input biomedical text is mapped as a graph where meaningful sentences are evaluated as the central node and the critical associations between them. The… More >

S. Vahdati¹, D. Mirzaei²

CMES-Computer Modeling in Engineering & Sciences, Vol.109-110, No.3, pp. 247-262, 2015, DOI:10.3970/cmes.2015.109.247

Abstract In this paper we present a meshless collocation method based on the moving least squares (MLS) approximation for numerical solution of the multiasset (d-dimensional) American option in financial mathematics. This problem is modeled by the Black-Scholes equation with moving boundary conditions. A penalty approach is applied to convert the original problem to one in a fixed domain. In finite parts, boundary conditions satisfy in associated (d-1)-dimensional Black-Scholes equations while in infinity they approach to zero. All equations are treated by the proposed meshless approximation method where the method of lines is employed for handling the More >

Sang-Ri Yi¹, Boyoung Kim², Jun Won Kang^2,3

CMES-Computer Modeling in Engineering & Sciences, Vol.106, No.2, pp. 77-104, 2015, DOI:10.3970/cmes.2015.106.077

Abstract This study is concerned with the development of new mixed unsplitfield perfectly matched layers (PMLs) for the simulation of plane electromagnetic waves in heterogeneous unbounded domains. To formulate the unsplit-field PML, a complex coordinate transformation is introduced to Maxwell's equations in the frequency domain. The transformed equations are converted back to the time domain via the inverse Fourier transform, to arrive at governing equations for transient electromagnetic waves within the PML-truncated computational domain. A mixed finite element method is used to solve the PML-endowed Maxwell equations. The developed PML method is relatively simple and straightforward More >

M. Dehghan¹, M. Abbaszadeh², A. Mohebbi³

CMES-Computer Modeling in Engineering & Sciences, Vol.100, No.5, pp. 399-444, 2014, DOI:10.3970/cmes.2014.100.399

Abstract In this paper three numerical techniques are proposed for solving the system of N-coupled nonlinear Schrödinger (CNLS) equations. Firstly, we obtain a time discrete scheme by approximating the first-order time derivative via the forward finite difference formula, then for obtaining a full discretization scheme, we use the Kansa’s approach to approximate the spatial derivatives via radial basis functions (RBFs) collocation methodology. We introduce the moving least squares (MLS) approximation and radial point interpolation method (RPIM) with their shape functions, separately. It should be noted that the shape functions of RPIM unlike the shape functions of the… More >

J. Sladek¹, V. Sladek¹, P. Stanak¹, P.H. Wen², S.N. Atluri³

CMES-Computer Modeling in Engineering & Sciences, Vol.85, No.6, pp. 543-572, 2012, DOI:10.3970/cmes.2012.085.543

Abstract A meshless local Petrov-Galerkin (MLPG) method is applied to solve laminate piezoelectric plates described by the Reissner-Mindlin theory. The piezoelectric layer can be used as a sensor or actuator. A pure mechanical load or electric potential are prescribed on the top of the laminated plate. Both stationary and transient dynamic loads are analyzed here. The bending moment, the shear force and normal force expressions are obtained by integration through the laminated plate for the considered constitutive equations in each lamina. Then, the original three-dimensional (3-D) thick plate problem is reduced to a two-dimensional (2-D) problem. More >

Zhen Luo¹, Nong Zhang¹, Tao Wu^2,3

CMES-Computer Modeling in Engineering & Sciences, Vol.85, No.4, pp. 299-328, 2012, DOI:10.3970/cmes.2012.085.299

Abstract This paper presents a meshless Galerkin level-set method (MGLSM) for shape and topology optimization of compliant mechanisms of geometrically nonlinear structures. The design boundary of the mechanism is implicitly described as the zero level set of a Lipschitz continuous level set function of higher dimension. The moving least square (MLS) approximation is used to construct the meshless shape functions with the global Galerkin weak-form in terms of a set of arbitrarily distributed nodes. The MLS shape function is first employed to parameterize the level set function via the surface fitting rather than interpolation, and then… More >

D. Ngo-Cong, N. Mai-Duy, W. Karunasena, T. Tran-Cong

CMES-Computer Modeling in Engineering & Sciences, Vol.83, No.5, pp. 459-498, 2012, DOI:10.3970/cmes.2012.083.459

Abstract The partition of unity method is employed to incorporate the moving least square (MLS) and one dimensional-integrated radial basis function (1D-IRBFN) techniques in a new approach, namely local MLS-1D-IRBFN or LMLS-1D-IRBFN. This approach leads to sparse system matrices and offers a high level of accuracy as in the case of 1D-IRBFN method. A new numerical procedure based on the 1D-IRBFN method and LMLS-1D-IRBFN approach is presented for a solution of fluid-structure interaction (FSI) problems. A combination of Chorin's method and pseudo-time subiterative technique is presented for a transient solution of 2-D incompressible viscous Navier-Stokes equations More >

Mehdi Tatari¹, Maryam Kamranian², Mehdi Dehghan²

CMES-Computer Modeling in Engineering & Sciences, Vol.82, No.1, pp. 1-28, 2011, DOI:10.32604/cmes.2011.082.001

Abstract In this paper, the finite point method (FPM) is presented for solving nonlinear reaction-diffusion systems which are often employed in mathematical modeling in developmental biology. In order to avoid directly solving a coupled nonlinear system, a predicator-corrector scheme is applied. The finite point method is a truly meshfree technique based on the combination of the moving least squares approximation on a cloud of points with the point collocation method to discretize the governing equations. The lack of dependence on a mesh or integration procedure is an important feature, which makes the FPM simple, efficient and More >