• 한국통신학회논문지
• /
• 제22권3호
• /
• pp.493-505
• /
• 1997

In this paper, a wavelet image compression method using human visually estimated quantizer is proposed. The quantizer has three components. These are constructed by using effects of frequency band, background luminance, and spatial masking. The first quantization factor is a fixed constant value for each band. The second factor is calculated by averaging four wavelet coefficients in the lowest frequency band. The third factor is determined by the difference between wavelet coefficients in the lowest frequency band. Arithmetic coding is used for encoding quantized wavelet coefficients. Coefficients in the lowest band are transmitted without loss. Therefore the compressed image is decompressed by using three quantization factors which can be calculated in the receiver. Compared with previous image compression methods which adopted human visual system, the proposed method shows improved results with less computational cost.

• 지식경영연구
• /
• 제6권2호
• /
• pp.105-118
• /
• 2005

The future will see all industries become technology-driven in the competitive global market place. Firms with deep technological roots and innovation strategies have some advantages. Business valuation of technology is critical to the future of firm's business. In this situation widely used scoring valuation is not enough to evaluate relative business competitiveness associated with technology and to assign its relative ranking category. Therefore, a more useful and comprehensive new valuation approach, which is called business composite index, is needed to complement and to enhance the existing scoring valuation approach. In this research, statistical factor analysis is applied to determine the common factors and to estimate associated weights. And business composite index, which is a kind of weighted scoring method, is derived based on the results of factor analysis. This research shows that business composite index is considered very useful to measure the business relative strength of individual technology and also to assign its relative ranking category instead of absolute ranking based on scoring valuation approach.

• International Journal of Contents
• /
• 제1권2호
• /
• pp.26-30
• /
• 2005

According to the fast growth of information on the Internet, it is becoming increasingly difficult to find and organize useful information. To reduce information overload, it needs to exploit automatic text classification for handling enormous documents. Support Vector Machine (SVM) is a model that is calculated as a weighted sum of kernel function outputs. This paper describes a document classifier for web documents in the fields of Information Technology and uses SVM to learn a model, which is constructed from the training sets and its representative terms. The basic idea is to exploit the representative terms meaning distribution in coherent thematic texts of each category by simple statistics methods. Vector-space model is applied to represent documents in the categories by using feature selection scheme based on TFiDF. We apply a category factor which represents effects in category of any term to the feature selection. Experiments show the results of categorization and the correlation of vector length.

• International journal of advanced smart convergence
• /
• 제12권2호
• /
• pp.17-24
• /
• 2023

Given the surge in the elderly population, and increasing in dementia cases, there is a growing interest in digital therapies that facilitate steady remote treatment. However, in the cognitive assessment of digital therapies through clinical trials, the absence of log data as an essential evaluation factor is a significant issue. To address this, we propose a solution of utilizing weighted derived variables based on high-importance variables' accuracy in log data utilization as an indirect cognitive assessment factor for digital therapies. We have validated the effectiveness of this approach using machine learning techniques such as XGBoost, LGBM, and CatBoost. Thus, we suggest the use of log data as a rapid and indirect cognitive evaluation factor for digital therapy users.

• 한국태양에너지학회 논문집
• /
• 제36권6호
• /
• pp.13-24
• /
• 2016

The government introduced a mandatory installation system of new & renewable energy for public building to meet the target of greenhouse gas reduction and also suggest a correction factor for new renewable energy to expand the installation of various new & renewable energy systems. The introduction of correction factors, however, was followed by the reduction of installation size of new & renewable energy sources. Assuming that it was caused by a correction factor for each new renewable energy source calculated by the initial costs, this study proposed a new correction factor approach based on payback periods to reflect the technology element in the calculation process of correction factors additionally. The application results of new correction factors show that it was possible to do complex calculations including the economic and technological aspects to select a new & renewable energy system and that the installation size was also enlarged.

• 한국정보과학회논문지:정보통신
• /
• 제37권5호
• /
• pp.409-414
• /
• 2010

무선 센서 네트워크에서 센서노드의 지리적인 위치를 요구하는 응용들이 현저하게 증가하고 있다. 최근 다양한 위치 측위 알고리즘들이 제안 되었지만, 대부분의 알고리즘은 특정한 하드웨어로 얻은 RSSI와 LQI 측정치를 기반으로 위치를 추정하고 있다. 본 논문에서는 이러한 추가적인 정보를 이용하지 않아도 기존 연구와 근사한 측정 결과를 얻을 수 있는 '가중 다중 링을 이용한 측위' 알고리즘 WMRL(Weighted Multiple Rings Localization)을 제안한다. 고정노드(anchor nodes)들이 배치되어 있으며, 각 고정노드는 주기적으로 서로 다른 신호 세기의 비콘(beacon) 신호를 송출한다고 가정한다. 그러면, 비콘 신호는 공간상에 링을 형성하게 되며, 파워 레벨의 세기에 따라 다수의 동심원을 형성하는 동시에 링 간에 교차영역을 생성한다. 본 논문에서는 효율적인 측위 계산을 위해 각 링의 거리 비율에 따른 가중치 모텔을 제안한다. 또한, 센서노드는 수신이 가능한 고정노드로부터 가장 가까운 링을 발견할 수 있으며, 이를 활용하여 센서노드는 자신의 위치를 고정노드 좌표의 가중 합으로 구한다. 제안된 알고리즘은 분산적으로 위치를 계산할 수 있으며, 추가적인 하드웨어를 요구하지 않는다. 추가적으로, 비 신뢰적인 RSSI 및 LQI에 의존하지 않고, 각 링 간의 거리 비율로 측위가 가능한 것이 특정이다. 그럼에도 불구하고, WMRL은 시뮬레이션 결과 2개의 링, 즉 2개의 파워 레벨로 구성하였을 경우에는 기존의 centroid 방식보다 평균 측위 에러가 2배 감소하였고, 3개의 링을 구성하였을 경우에는 WCL(Weighted Centroid Localization)과 대등한 측위 결과를 보였다.

• 비파괴검사학회지
• /
• 제22권5호
• /
• pp.465-473
• /
• 2002

영상 데이타는 전송, 검출 및 처리과정에서 여러 잡음에 의해 훼손될 수 있다. 적응성 가중 메디안 필터라는 공간변화 필터를 사용하여 방사선 투과영상의 양자 잡음을 제거하였다. 제안된 필터는 처리 윈도우 내 각 픽셀의 국소 통계치의 변화에 따라 필터의 성능이 변화하여 에지를 최대한 보존하면서 잡음만을 제거하고자 이러한 국소 통계 값에 근거한 적응성 가중 메디안 휠터 (AWMF)를 제시한다. AWMF를 구현함에 있어 두 가지 방법으로 나뉘는데, 우선 국소 통계의 특성에 따라 3개의 영역으로 분류하여 가중치를 부여하는 간단한 비선형 필터이고, 다음으로는 잡음모델로부터 국소 통계의 특성에 따라 경계(edge) 영역과 균일 영역으로 구분하여 영상시스템에 적당한 균일 척도 값을 구하여 영상의 공간적인 변화 정도를 판단하는 기준이 되도록 하였다. 제안한 알고리듬은 IBM-PC 상에서 C 언어로 구현하였으며 AWMF가 다른 잡음 제거 필터들과의 성능 비교를 통하여 경계는 보존하면서 잡음은 최대한 제거하는 우수한 필터임을 검증하였다.

• Communications for Statistical Applications and Methods
• /
• 제15권6호
• /
• pp.825-835
• /
• 2008

계량경제학 분야에서 널리 쓰이는 MGARCH(multivariate GARCH)모형은 여러개의 시계열자료들의 변동성을 함께 모형화한다. 그러나 변수가 많아질수록 추정해야 할 모수의 수가 급격하게 늘어나는 문제점이 있다. 본 연구에서는 인자 모형을 통해 자료의 차원을 축소시킴로써 이러한 문제를 해결하고자 하였다. 국내의 주가수익률 자료에 통계적 인자 모형과 fundamental factor model을 적용하여 각각의 의미 있는 인자들을 얻은 후 이를 MGARCH모형에 적합시켰다. 또한 두 인자모형을 바탕으로 얻어진 최종 모형들의 MSE, MAD와 VaR(Value at Risk)를 계산하여 예측력을 비교하고자 한다.

• 영어어문교육
• /
• 제17권1호
• /
• pp.191-208
• /
• 2011

The quantitative study investigated perceptions of evaluative criteria in L2 writing between two groups - learners (N=212) and instructors (N=52) in Korea. Specifically, the purpose of the study is (1) to examine learners' and instructors' perceptions on evaluative criteria in L2 writing and to provide empirical evidence concerning how they respond to a list of them and (2) to ultimately devise appropriate rating criteria applicable to an EFL context like Korea. Analyses of evaluative criteria were conducted using factor analysis and yielded the following results: learner and instructor groups perceived the evaluative criteria differently and weighted them in a different way. For the learner group, the combined elements of grammar and language in use were identified as Factor 1 and mechanics as Factor 2. The results may infer that learners' response patterns are primarily linked to their instructors' writing practice in class, which may largely focus on grammatical knowledge based on lexical use and mechanical accuracy. Similarly, the instructor group acknowledged grammatical knowledge as Factor 1 and lexical use as Factor 2. The first two factors found in both learner and instructor groups indicate that in an EFL context like Korea, the form-then-content way of teaching and learning is still being considered more effective in L2 writing than any other method. Taking into consideration these perceptive similarities and differences between learners and instructors, the categories of evaluative criteria in writing include content and organization, grammar, mechanics, language in use, and flow of the essay, respectively.

• 응용통계연구
• /
• 제25권5호
• /
• pp.829-835
• /
• 2012

When the input features are generated by factors in a classification problem, it is more meaningful to identify important factors, rather than individual features. The $F_{\infty}$-norm support vector machine(SVM) has been developed to perform automatic factor selection in classification. However, the $F_{\infty}$-norm SVM may suffer from estimation inefficiency and model selection inconsistency because it applies the same amount of shrinkage to each factor without assessing its relative importance. To overcome such a limitation, we propose the adaptive $F_{\infty}$-norm ($AF_{\infty}$-norm) SVM, which penalizes the empirical hinge loss by the sum of the adaptively weighted factor-wise $L_{\infty}$-norm penalty. The $AF_{\infty}$-norm SVM computes the weights by the 2-norm SVM estimator and can be formulated as a linear programming(LP) problem which is similar to the one of the $F_{\infty}$-norm SVM. The simulation studies show that the proposed $AF_{\infty}$-norm SVM improves upon the $F_{\infty}$-norm SVM in terms of classification accuracy and factor selection performance.