• The KIPS Transactions:PartB
• /
• v.9B no.6
• /
• pp.735-744
• /
• 2002

This paper proposes learning performance improvement of support vector machine using the kernel relaxation and the dynamic momentum. The dynamic momentum is reflected to different momentum according to current state. While static momentum is equally influenced on the whole, the proposed dynamic momentum algorithm can control to the convergence rate and performance according to the change of the dynamic momentum by training. The proposed algorithm has been applied to the kernel relaxation as the new sequential learning method of support vector machine presented recently. The proposed algorithm has been applied to the SONAR data which is used to the standard classification problems for evaluating neural network. The simulation results of proposed algorithm have better the convergence rate and performance than those using kernel relaxation and static momentum, respectively.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.6 no.4
• /
• pp.89-99
• /
• 2003

Smoothed particle hydrodynamics, SPH, is a gridless Lagrangian technique which is a useful alternative numerical analysis method to simulate high velocity deformation problems as well as astrophysical and cosmological problems. The SPH method brings about some difficulties such as tensile Instability and stress oscillation. A new SPH method, so called normalized algorithm, was introduced to overcome these difficulties. In this paper we aimed to estimate this method and have developed an one-dimensional normalized SPH program. The high velocity impact model of an aluminum bar has been analysed by using the developed program and a commercial hydrocode, LS-DYNA. The obtained numerical results showed good agreement with the results of the same model in reference. The program also showed more stable results than those of LS-DYNA in stress oscillation. We hopefully expect that the developed one-dimensional normalized SPH program can be used to solve hydrodynamic problems especially for explosive detonation analysis.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.76-78
• /
• 2014

Naive Bayes (NB, for shortly) learning is more popular, faster and effective supervised learning method to handle the labeled datasets especially in which have some noises, NB learning also has well performance. However, the conditional independent assumption of NB learning imposes some restriction on the property of handling data of real world. Some researchers proposed lots of methods to relax NB assumption, those methods also include attribute weighting, kernel density estimating. In this paper, we propose a novel approach called NB Based on Attribute Weighting in Kernel Density Estimation (NBAWKDE) to improve the NB learning classification ability via combining kernel density estimation and attribute weighting.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.86-88
• /
• 2014

Naive Bayes (NB) assumption has some harmful effects in classification to the real world data. To relax this assumption, we now propose approach called Naive Bayes Mutual Information Attribute Weighting with Smooth Kernel Density Estimation (NBMIKDE) that combine the smooth kernel for attribute and attribute weighting method based on mutual information measure.

• Environmental and Resource Economics Review
• /
• v.18 no.1
• /
• pp.105-138
• /
• 2009

This study investigates the spatial distribution of the degree of water pollutants in Han River using data obtained by the water pollution observation stations. This study estimates a non -parametric kernel density function for each water pollutants, and tests a significant difference between two estimated distribution functions. Next, Generalized Entropy inequality indices are evaluated and this research tests difference of inequality indices between two years using bootstrapping method. Lastly in a dynamic of view, it is analyzed that there are significant changes in the ranking of water pollution level. Estimation results show that the degree of inequality in spatial distribution of water pollution tends to be stable or decreasing for last 15 years in a great part of water pollutants, and ranking of water pollution level hardly changes in Han River.

• Environmental and Resource Economics Review
• /
• v.13 no.1
• /
• pp.83-117
• /
• 2004

This study investigates the spatial distributions of the ambient levels of air pollutants ($SO_2$, $NO_2$, $O_3$, CO, and PM) in Seoul metropolitan area using the data obtained by the air pollution observation stations. This study estimated a non-parametric kernel density function and two types of inequality indices, Gini and Entropy. Our estimation results show that the degree of inequality in spatial distribution of air pollution, in general, tends to be stable or slightly decreasing for the period of 1990~2001. In addition, we found that there are significant dynamics of air pollution levels in terms of spatial ranking.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.7
• /
• pp.37-45
• /
• 2013

Context dependent fusion (CDF) is a fusion algorithm that combines multiple outputs from different classifiers to achieve better performance. CDF tries to divide the problem context into several homogeneous sub-contexts and to fuse data locally with respect to each sub-context. CDF showed better performance than existing methods, however, it is sensitive to noise due to the large number of parameters optimized and the innate linearity limits the application of CDF. In this paper, a variant of CDF using support vector machines (SVMs) for fusion and kernel principal component analysis (K-PCA) for context extraction is proposed to solve the problems in CDF, named CDF-SVM. Kernel PCA can shape irregular clusters including elliptical ones through the non-linear kernel transformation and SVM can draw a non-linear decision boundary. Regularization terms is also included in the objective function of CDF-SVM to mitigate the noise sensitivity in CDF. CDF-SVM showed better performance than CDF and its variants, which is demonstrated through the experiments with a landmine data set.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.4
• /
• pp.821-834
• /
• 2019

As cyber security threats increase, there is a growing demand for highly reliable systems. Common Criteria, an international standard for evaluating information security products, requires formal specification and verification of the system to ensure a high level of security, and more and more cases are being observed. In this paper, we propose highly reliable drone systems that ensure high level security level and trust. Based on the results, we use formal methods especially Z/EVES to improve the system model in terms of scheduling in the system kernel.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.167-183
• /
• 2023

In the recent field of recommendation systems, various studies have been conducted to model sparse data effectively. Among these, GLocal-K(Global and Local Kernels for Recommender Systems) is a research endeavor combining global and local kernels to provide personalized recommendations by considering global data patterns and individual user characteristics. However, due to its utilization of kernel tricks, GLocal-K exhibits diminished performance on highly sparse data and struggles to offer recommendations for new users or items due to the absence of side information. In this paper, to address these limitations of GLocal-K, we propose the GEase-K (Global and EASE kernels for Recommender Systems) model, incorporating the EASE(Embarrassingly Shallow Autoencoders for Sparse Data) model and leveraging side information. Initially, we substitute EASE for the local kernel in GLocal-K to enhance recommendation performance on highly sparse data. EASE, functioning as a simple linear operational structure, is an autoencoder that performs highly on extremely sparse data through regularization and learning item similarity. Additionally, we utilize side information to alleviate the cold-start problem. We enhance the understanding of user-item similarities by employing a conditional autoencoder structure during the training process to incorporate side information. In conclusion, GEase-K demonstrates resilience in highly sparse data and cold-start situations by combining linear and nonlinear structures and utilizing side information. Experimental results show that GEase-K outperforms GLocal-K based on the RMSE and MAE metrics on the highly sparse GoodReads and ModCloth datasets. Furthermore, in cold-start experiments divided into four groups using the GoodReads and ModCloth datasets, GEase-K denotes superior performance compared to GLocal-K.

• Korean Journal of Remote Sensing
• /
• v.27 no.5
• /
• pp.573-586
• /
• 2011

The MTF(modulation transfer function) is one of parameters to evaluate the performance of imaging systems. Also, it can be used to restore information that is lost by a harsh space environment (radioactivity, extreme cold/heat condition and electromagnetic field etc.), atmospheric effects and falloff of system performance etc. This paper evaluated the MTF values of images taken by DubaiSat-1 satellite which was launched in 2009 by EIAST(Emirates Institute for Advanced Science and Technology) and Satrec Initiative. Generally, the MTF was assessed using various methods such as a point source method and a knife-edge method. This paper used the slanted-edge method. The slantededge method is the ISO 12233 standard for the MTF measurement of electronic still-picture cameras. The method is adapted to estimate the MTF values of line-scanning telescopes. After assessing the MTF, we performed the MTF compensation by generating a MTF convolution kernel based on the PSF(point spread function) with image denoising to enhance the image quality.