Framework Of LDA Research Articles

Consumer Feedback Analysis Using LDA Approach in Cross-border E-commerce

Purpose - The objective of this research is to gain insight into consumer feedback regarding cross-border e-commerce, which is a pivotal avenue for companies venturing into global markets. It seeks to investigate the variety of themes present in consumer feedback in order to gain a more profound understanding of their preferences and concerns, ultimately influencing future market strategies. Design/Methodology – The present study utilizes an extensive collection of textual online evaluations and a text-mining LDA framework to analyze and classify potential themes in consumer feedback pertaining to cross-border e-commerce platforms. By quantifying the sentiment tendencies in consumer-related comments, sentiment analysis models that rely on emotion dictionaries as a foundation identify the polarity and intensity of emotions. Findings – The present study examined 1,500 consumer evaluations pertaining to Korean cosmetic products, sourced from a cross-border platform in Southeast Asia. The analysis unveiled five key subjects frequently referenced in consumer feedback. A subsequent sentiment analysis revealed that a significant proportion of consumer sentiments were positive. Nevertheless, consumers identified certain concerns in the minority of negative reviews, including packaging and delivery, product quality, and the absence of freebies. Originality/value - In a practical sense, this provides significant ramifications for the optimization of consumer experiences and the development of successful market strategies in the ever-changing global marketplace. This is especially advantageous for organizations striving for success in the fiercely competitive cross-border e-commerce industry. This study exceeds conventional survey methods in its methodology by offering more comprehensive and datacentric understandings of consumer attitudes and behaviors.

Dimensionality reduction by collaborative preserving Fisher discriminant analysis

Sparse representation-based classifier (SRC) and collaborative representation-based classifier (CRC) are two commonly used classifiers. There has been pointed out that the utilization of all the training samples in representing a query sample (i.e. the least square part), which reflects the collaborative representation mechanism of SRC and CRC, is more important than the norm constraint on the coding coefficients for classification. From this perspective, both SRC and CRC can be viewed as collaborative representation (CR) but with different norm (i.e. L1 and L2) constraints on the coding coefficients. In this paper, two collaborative preserving Fisher discriminant analysis approaches are proposed for linear dimensionality reduction, in which both the local geometric information hidden in the CR coefficients and the global discriminant information inherited from Fisher/linear discriminant analysis (FDA/LDA) are effectively fused. Specifically, a datum adaptive graph is first built via CR with L1 or L2 norm constraint (corresponding to L1CPFDA and L2CPFDA, respectively), and then incorporated into the LDA framework to seek a powerful projection subspace with analytic solution. Both theoretical and experimental analysis of L1CPFDA and L2CPFDA show that they can best preserve the collaborative reconstruction relationship of the data and discriminate samples of different classes as well. Moreover, LDA is a special case of L1CPFDA and L2CPFDA and the available number of projection directions of them are twice that of LDA empirically. Experimental results on ORL, AR and FERET face databases and COIL-20 object database demonstrate their effectiveness, especially in low dimensions and small training sample size.

Double correlation model for operational risk: Evidence from Chinese commercial banks

The Basel Accord requires commercial banks to meet a capital requirement for operational risk. Therefore, a reliable operational risk measurement is of great significance for financial institutions. In this paper, based on the framework of LDA, we propose a more appropriate model (Double Correlation model) in correlation structure construction, which considers both frequency correlation and mean severity correlation simultaneously. On this basis, annual loss scenarios for each risk cell are restored and VaR is also estimated by Monte Carlo simulation. Empirical study adopts the data of Chinese commercial banks disclosed to the public and contains the comparison of double correlation model and the other four correlation models. The result indicates that the VaR estimated by the double correlation model is significantly lower than that upon the assumption of comonotonicity, proving the existence of risk diversification. Meanwhile, the VaR estimated by double correlation model is significantly lower than that of the other two correlation models when confidence level is lower than 99% but the relationships between them will reverse if the confidence level rises to more than 99%.

Algorithms of Two-Dimensional Projection of Digital Images in Eigensubspace: History of Development, Implementation and Application

Algorithms for projection of digital images into their eigensubspaces in the framework of linear methods PCA, LDA, PLS and CCA are considered. The history of these methods development of over the past 100 years is given against the backdrop of the emergence of new areas of their application and changing requirements in relation to them. It is shown that this development was initiated by four basic requirements stemming from modern tasks and practice of digital image processing and, first of all, face images (FI). The first requirement is the use of PCA, LDA, PLS and CCA methods in conditions of both a small and extremely large samples of ILs in the initial sets. The second requirement is related to the criterion that determines its eigenbasis, and which should provide, for example, the minimum error of FI approximation, the improvement of clustering in its eigensubspace or the maximum correlation (covariance) between data sets in the subspace. The third one is related to the possibility of applying the methods under consideration to the tasks of processing two or more sets of images from different sensors or several sets of any number matrices. These three requirements led to the emergence, development and application of methods of two-dimensional projection into their eigensubspaces–2DPCA, 2DLDA, 2DPLS and 2DCCA. Several basic branches of algorithmic implementation of these methods are considered (iterative, not iterative, based on SVD, etc.), their advantages and disadvantages are evaluated, and examples of their use in practice are also shown. Finally, the fourth requirement is the possibility of realizing two-dimensional projections of FI (or other numerical matrices) directly in the layers of convolutional neural networks (CNN/Deep NN) and/or integrating their functions into NN by separate blocks. The requirement and examples of its solution are discussed. Estimates of computational complexity for the presented algorithms and examples of solving specific problems of image processing are given.

The GW electronic structure of cubic RbMF3 perovskites (M = Be, Mg, Ca, Sr, Ba)

The electronic energy band spectra of cubic RbMF3 perovskites (M = Be, Mg, Ca, Sr, Ba) have been evaluated within the projector augmented waves (PAW) approach by means of the ABINIT code. The Kohn-Sham single-particle states have been found in the LDA framework. Our parameters of the electron energy bands obtained in the LDA are in good agreement with the published results of other authors. The calculated GW band gaps and dielectric constants are well compared with the available experimental data. The quasiparticle energies and density of states of electrons as well as the dielectric constants were obtained in the approximation GW for the first time.

Anomalous band-gap bowing of AlN1−xPx alloy

Electronic structure of zinc blende AlN1−xPx alloy has been calculated from first principles. Structural optimization has been performed within the framework of LDA and the band-gaps calculated with the modified Becke–Jonson (MBJLDA) method. Two approaches have been examined: the virtual crystal approximation (VCA) and the supercell-based calculations (SC). The composition dependence of the lattice parameter obtained from the SC obeys Vegard’s law whereas the volume optimization in the VCA leads to an anomalous bowing of the lattice constant. A strong correlation between the band-gaps and the structural parameter in the VCA method has been observed. On the other hand, in the SC method the supercell size and atoms arrangement (clustered vs. uniform) appear to have a great influence on the computed band-gaps. In particular, an anomalously big band-gap bowing has been found in the case of a clustered configuration with relaxed geometry. Based on the performed tests and obtained results some general features of MBJLDA are discussed and its performance for similar systems predicted.

Band structure, optical properties and infrared spectrum of glycine–sodium nitrate crystal

Glycine–sodium nitrate, GSN, crystals were grown from a stoichiometric solution by slow cooling technique and were characterized by optical absorption and FTIR spectroscopy. The data collected by FTIR were compared with the vibrational spectrum theoretically obtained by using DMol code in the local density approximation LDA. Moreover, the crystal band structure, the density of states, and the optical absorption data were calculated by using the CASTEP code within the framework of LDA and the generalized gradient approximation GGA. The calculations are in good agreement with the structure and properties of GSN; e.g., the optical transparency in visible region, the low density, the insulate character, and the bipolar form of glycine molecule.

Experimental and theoretical studies of LiNi1/3Mn1/3M1/3O2 [M=Mo and Rh] for cathode material

First principle calculation using VASP in the frame work of LDA has been performed for cation substitution of 4d transition metals M (Mo and Rh) in LiNi1/3Mn1/3M1/3O2 in the α-NaFeO2 structure. The intercalation potentials and relative formation energies of the fully lithiated LiNi1/3Mn1/3M1/3O2, fully delithiated Ni1/3Mn1/3M1/3O2, and the intermediate phases Li1/3Ni1/3Mn1/3M1/3O2 and Li2/3Ni1/3Mn1/3M1/3O2 were computed by employing a generalized alloy theory. A minute substitution of cationic Mo in LiNiMnO2 was experimentally investigated to examine the effect of the Mo substitution on the electrochemical properties. Li [Ni0.5+xMn0.5–2xMox]O2 (x=0.02) were synthesized by sol–gel method as positive cathode material for Li ion microbatteries. X-ray diffraction pattern of the powder material suggests the formation of the compound in α-NaFeO2 layered structure. CV data indicates no other electrochemical active phase in the voltage range under study.

Effect of substitutions and defects in half-HeuslerFeVSbstudied by electron transport measurements and KKR-CPA electronic structure calculations

The structural and electron transport properties of the pure and $\mathrm{Co}\text{\ensuremath{-}}$, $\mathrm{Ti}\text{\ensuremath{-}},$ and $\mathrm{Zr}\text{\ensuremath{-}}$substituted $\mathrm{FeVSb}$ half-Heusler phases have been investigated using x-ray diffraction, M\ossbauer spectroscopy, and Electron Probe Microscopy Analysis as well as resistivity, thermopower, and Hall effect measurements in the $80--900\phantom{\rule{0.3em}{0ex}}\mathrm{K}$ temperature range. In a parallel study, the electronic structures of $\mathrm{FeVSb}$ and the aforementioned alloys were calculated using the Korringa-Kohn-Rostoker method with the coherent potential approximation (KKR-CPA) in the LDA framework. The electronic densities of states and dispersion curves were obtained. The crystal structure stability and site preference analysis were addressed using total energy computations. Most of these experimental results correspond to electronic structure computations only if they take into account extra crystal defects such as antisite defects or vacancies present to various extents in the samples. Indeed a remarkable variation of KKR-CPA density of states occurring both in $\mathrm{FeVSb}$ and ${\mathrm{FeV}}_{1\ensuremath{-}x}{\mathrm{Zr}}_{x}\mathrm{Sb}$ including defects may explain why $\mathrm{FeVSb}$ is not fully semiconducting as well as why there is a change of the thermopower sign in the ${\mathrm{FeV}}_{1\ensuremath{-}x}{\mathrm{Zr}}_{x}\mathrm{Sb}$ versus $\mathrm{x}$ content.

Effect of short-range magnetic ordering on electron energy-loss spectra in spinels

The energy-loss near-edge structure (ELNES) at the oxygen $K$ edge in two $AB_2O_4$ spinels (A=Mg, B=Al,Cr) is reported. In $MgAl_2O_4$ the experimental data is successfully modelled within the LDA framework. In the case of $MgCr_2O_4$ spin-polarisation, in the form of antiferromagnetic ordering on the Cr sub-lattice, must be included despite the fact that the measurements were performed at approximately 30 times above $T_N$. A model in which dynamic short-range antiferromagnetic ordering is present at room temperature is proposed to explain the results of the experiments and calculations.

Thermoelectric Properties and Electronic Structure of CoSb3 Doped with Se and Te.

AbstractFor Abstract see ChemInform Abstract in Full Text.

Thermoelectric properties and electronic structure of CoSb 3 doped with Se and Te

This work presents structural and electron transport properties of Se- and Te-doped CoSb 3 skutterudites in relation to results of electronic structure calculations performed for this system. The physical quantities were determined from crystal structure studies and measurements of transport coefficients: electrical conductivity, thermal conductivity and Seebeck coefficient as a function of composition and temperature. The electronic structure of CoSb 3 was calculated using the Korringa–Kohn–Rostoker method with the LDA framework. Dispersion curves E( k) in CoSb 3 computed for experimental crystallographic parameters exhibit a very narrow indirect energy gap at the Fermi level, which however strongly varies along different k-directions. The KKR valence and conduction band curvatures reflect effective masses estimated from experimental data well. The effect of doping CoSb 3 with Se and with Te as well as the dopants’ influence on density of electronic states was investigated by the KKR method combined with the coherent potential approximation (CPA). This method was also used to show an impact of various crystal defects on electronic structure in pure Se- and Te-doped CoSb 3.

Electronic phase diagram of the XTZ (X=Fe, Co, Ni; T=Ti, V, Zr, Nb, Mn; Z=Sn, Sb) semi-Heusler compounds

Abstract Some general aspects of electronic structure of the XTZ semi-Heusler systems are discussed using the Korringa–Kohn–Rostoker computations within the LDA framework. An energy gap appears for a total number of 18 valence electrons, which corresponds to semiconductors. This gap is still maintained in the density of states in other XTZ compounds with different electron concentration (EC). This peculiar electronic structure gives rise to a variety of different physical properties, starting from EC=16 (FeTiSn) and EC=17 (FeTiSb, CoTiSn) (paramagnetic or ferromagnetic) metals, through semiconductors (NiTiSn, NiZrSn, NiYSb, FeVSb, CoTiSb, etc.) and EC=19 (CoVSb, CoNbSb, NiTiSb) metallic phases. The reason for the instability of the semi-Heusler phase of NiVSb (EC=20) is discussed using the total energy KKR analysis. The characteristic gap in the density of states is also maintained for the minority spin projection in the case of the half-metallic ferromagnetism in CoMnSb (EC=21) and NiMnSb (EC=22). Some of the semiconducting compounds (NiTiSn, FeVSb, NiYSb) seem to possess promising thermoelectrical properties.

Electron momentum-space densities and Fermi surface ofLi100−xMgx(0<~x<~40)alloys: Compton scattering experiment versus theory

Directional Compton profiles (CP's) of ${\mathrm{Li}}_{100\ensuremath{-}x}{\mathrm{Mg}}_{x}(0<~x<~40)$ were measured for (11)12 directions of the momentum transfer q with (0.14) 0.16 a.u. $(\mathrm{a}.\mathrm{u}.=\mathrm{a}\mathrm{t}\mathrm{o}\mathrm{m}\mathrm{i}\mathrm{c}$ units: $\ensuremath{\Elzxh}=e=m=1)$ momentum space resolution using synchrotron x rays, monochromatized to (31) 59 keV (values in brackets refer to $x=0).$ The total experimental valence electron CP's, their second derivatives, and the associated directional differences were compared with the corresponding Korringa-Kohn-Rostoker--coherent-potential-approximation--local-density-approximation (KKR--CPA--LDA) computations. Discrepancies between measurements and calculations are partly traced back to an inadequate treatment of electron correlation effects within the LDA framework. The three-dimensional (3D) valence-electron momentum density, \ensuremath{\rho}(p), as well as the 3D occupation number density $N(\mathbf{k}),$ were reconstructed from experimental data using the Fourier-Bessel method. The reconstructed \ensuremath{\rho}(p)'s exhibit evidence of higher momentum components due to 110 umklapp processes, whose variations with x are in reasonable agreement with the KKR-CPA calculations. The experimental $N(\mathbf{k})$'s provide a direct measure of the Fermi-surface anisotropy and show the presence of necks on the 110 Brillouin-zone boundaries for $x>~13$ in good accord with 2D positron annihilation experiments.

Electronic and optical properties of semiconducting iron disilicide

Calculations of electronic properties of β-FeSi 2 by the LMTO method within the strict LDA scheme with correlation and exchange effects included have revealed new interesting details of the band structure of this material. We conclude that β-FeSi 2 is a quasi-direct semiconductor. On the basis of direct extended comparison of theoretical and experimental interband optical spectra of semiconducting iron disilicide, one may conclude on the correct description of the electronic and optical properties in the framework of LDA in the extended energy range up to 5 eV below and above the Fermi energy. Anisotropy effects have been theoretically predicted to occur in the optical properties of β-FeSi 2.

Ab-initio adiabatic noble gas–metal interaction: the role of the induced polarization charge

Within an ab-initio DFT–LDA framework, we have investigated the polarization induced on He and Ne atoms by a jellium (Al-like) surface. It is shown that the displaced charge of the gas–metal system is spherically distributed around Ne. It is polarized instead for He with a dipole moment having the negative lobe towards the jellium surface. A relationship between the two different polarizations of He and Ne with corrugating and anti-corrugating behavior is discussed.