• Journal of the Korean Institute of Intelligent Systems
• /
• v.20 no.4
• /
• pp.534-540
• /
• 2010

High dimensionality and highly correlated features are the major characteristics of hyperspectral data. Linear projections such as LDA and its variants have been used in extracting low-dimensional features from high-dimensional spectral data. Regularization of LDA has been introduced to alleviate the overfitting that often occurs in a small-sized training data set and leads to poor generalization performance. Among them, a smoothness regularized LDA seems to be effective in the feature extraction for hyperspectral data due to its capability of utilizing the high correlatedness. This paper studies the performance of the regularized LDA in hyperspectral data classification experimentally with varying conditions of the training data. In addition, a new dual smoothness regularized LDA is proposed and evaluated that makes use of both the spectral-domain and spatial-domain correlations between neighboring pixels.

• The KIPS Transactions:PartD
• /
• v.13D no.7 s.110
• /
• pp.909-922
• /
• 2006

The core assets include all properties which consist of an application in Product Line Engineering. The requirement, one of the core assets, is a basis of other core assets and commonality and variability of other core assets are classified by the requirement. accordingly, commonality and variability of the domain requirement should be managed objectively and it is necessary to make a process to reuse the domain requirements. However the requirement is analyzed by domain experts or developers without proper process. In this paper, we proposed the 4 activities: (1)the domain scoping, (2)the extraction and generalization of the domain requirement, (3)the domain requirement analyzing and modeling, (4)the change management, and sub activities. For all reasons given previously, it is possible to reduce the development time and cost by reusing the architectures and components related to the domain requirement. In addition, it is possible to increase the quality of the artifacts produced based on the requirements by managing them systematically.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.504-509
• /
• 2008

Support Vector Machine (SVM) is one of powerful learning machine and has been applied to varying task with generally acceptable performance. The success of SVM for classification tasks in one domain is affected by features which represent the instance of specific class. Given the representative and discriminative features, SVM learning will give good generalization and consequently we can obtain good classifier. In this paper, we will assess the problem of feature choices for human detection tasks and measure the performance of each feature. Here we will consider HOG-family feature. As a natural extension of SVM, we combine SVM with Principal Component Analysis (PCA) to reduce dimension of features while retaining most of discriminative feature vectors.

• Bulletin of the Korean Mathematical Society
• /
• v.51 no.2
• /
• pp.409-421
• /
• 2014

This paper deals with the superlinear elliptic problem without Ambrosetti and Rabinowitz type growth condition of the form: $$\{-div\((1+\frac{|{\nabla}u|^{p(x)}}{\sqrt{1+|{\nabla}u|^{2p(x)}}}})|{\nabla}u|^{p(x)-2}{\nabla}u\)={\lambda}f(x,u)\;a.e.\;in\;{\Omega}\\u=0,\;on\;{\partial}{\Omega}$$ where ${\Omega}{\subset}R^N$ is a bounded domain with smooth boundary ${\partial}{\Omega}$, ${\lambda}$ > 0 is a parameter. The purpose of this paper is to obtain the existence results of nontrivial solutions for every parameter ${\lambda}$. Firstly, by using the mountain pass theorem a nontrivial solution is constructed for almost every parameter ${\lambda}$ > 0. Then we consider the continuation of the solutions. Our results are a generalization of that of Manuela Rodrigues.

• Asia pacific journal of information systems
• /
• v.2 no.1
• /
• pp.35-56
• /
• 1992

This paper proposes a synergism of neural networks (NN) and knowledge-based decision support system (KBDSS) to effectively design an intelligent strategic planning system. Since conventional KBDSS becomes inoperative partially or totally when problem deviates slightly from the expected problem-domain, a new DSS concept is needed for designing an effective strategic planning system, where strategic planning environment is usually turbulent and consistently changing. In line with this idea, this paper developes a NN-based DSS, named ConDSS, incorporating the generalization property of NN into its knowledge base. The proposed ConDSS was extensively operated in an experimentally designed environment with three models: BCG matrix, Growth/Gain matrix, and GE matrix. The results proved very promising when encountered with unforeseen situations in comparisons with conventional KBDSS.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.29 no.5
• /
• pp.421-428
• /
• 2016

The present work extends the recent variational formulation to more general time-dependent problems. Thus, based upon recent works of variational formulation in dynamics and pure heat diffusion in the context of the extended framework of Hamilton's principle, formulation for fully coupled thermoelasticity is developed first, then, with thermoelasticity-poroelasticity analogy, poroelasticity formulation is provided. For each case, energy conservation and energy dissipation properties are discussed in Fourier transform domain.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2010.05a
• /
• pp.74-75
• /
• 2010

Inflow broadband noise is generated when turbulence in the rotor wakes impinges on the downstream stator vanes. In this paper a three-dimensional model is developed to investigate the broadband noise due to turbulence-cascade interaction. In the newly-developed model, we consider the effects of incident turbulent gust component in span-wise direction on the inflow broadband noise. The quasi-three-dimensional theory is deduced based on the tonal analytic theory of Smith (1972) and two-dimensional broadband noise generalization by Cheong et al. (2006; 2009). Extending the modified LINSUB code, quasi-three-dimensional computational results are presented. Finally, we compare these computational results with time-domain results to validate the theory.

• Bulletin of the Korean Mathematical Society
• /
• v.54 no.5
• /
• pp.1757-1771
• /
• 2017

Let K be a field and G be a group of its automorphisms. It follows from Speiser's generalization of Hilbert's Theorem 90, [10] that any K-semilinear representation of the group G is isomorphic to a direct sum of copies of K, if G is finite. In this note three examples of pairs (K, G) are presented such that certain irreducible K-semilinear representations of G admit a simple description: (i) with precompact G, (ii) K is a field of rational functions and G permutes the variables, (iii) K is a universal domain over field of characteristic zero and G its automorphism group. The example (iii) is new and it generalizes the principal result of [7].

• Kyungpook Mathematical Journal
• /
• v.58 no.2
• /
• pp.203-219
• /
• 2018

This paper is devoted to the study of strongly ${\alpha}-semicommutative$ rings, a generalization of strongly semicommutative and ${\alpha}-rigid$ rings. Although the n-by-n upper triangular matrix ring over any ring with identity is not strongly ${\bar{\alpha}}-semicommutative$ for $n{\geq}2$, we show that a special subring of the upper triangular matrix ring over a reduced ring is strongly ${\bar{\alpha}}-semicommutative$ under some additional conditions. Moreover, it is shown that if R is strongly ${\alpha}-semicommutative$ with ${\alpha}(1)=1$ and S is a domain, then the Dorroh extension D of R by S is strongly ${\bar{\alpha}}-semicommutative$.

• Honam Mathematical Journal
• /
• v.31 no.4
• /
• pp.591-599
• /
• 2009

For given ${\alpha},{\omega}\;{\in}\;{\mathbb{R}}$ and ${\beta}$ > 1, let $T_{{\beta},{\alpha},{\omega}}$ be the skew-product transformation on the torus, [0, 1) ${\times}$ [0, 1) defined by (x, y) ${\longmapsto}\;({\beta}x,y+{\alpha}x+{\omega})$ (mod 1). In this paper, we give a criterion of ergodicity and weakly mixing for the transformation $T_{{\beta},{\alpha},{\omega}}$ when the natural extension of the given ${\beta}$-transformation can be viewed as a generalized baker's transformation, i.e., they flatten and stretch and then cut and stack a two-dimensional domain. This is a generalization of theorems in [10].