• Geophysics and Geophysical Exploration
• /
• v.10 no.2
• /
• pp.147-153
• /
• 2007

The conjugate gradient (CG) method is one of the most efficient algorithms for solving a linear system of equations. In addition to being used as a linear equation solver, it can be applied to a least-squares problem. When the CG method is applied to large-scale three-dimensional inversion of magnetotelluric data, two approaches have been pursued; one is the linear CG inversion in which each step of the Gauss-Newton iteration is incompletely solved using a truncated CG technique, and the other is referred to as the nonlinear CG inversion in which CG is directly applied to the minimization of objective functional for a nonlinear inverse problem. In each procedure we only need to compute the effect of the sensitivity matrix or its transpose multiplying an arbitrary vector, significantly reducing the computational requirements needed to do large-scale inversion.

• The Journal of the Acoustical Society of Korea
• /
• v.36 no.5
• /
• pp.338-344
• /
• 2017

In this paper, an interpolation method of HRTF (Head-Related Transfer Function) is proposed for small-sized measurement data set, especially. The proposed algorithm is based on acoustic modeling of HRTFs, and the algorithm tries to interpolate the HRTFs via estimation the model coefficients. However, the estimation of the model coefficients is hard if there is lack of measurement points, so the algorithm solves the problem by a data augmentation using the VBAP (Vector Based Amplitude Panning). Therefore, the proposed algorithm consists of two steps, which are data augmentation step based on VBAP and model coefficients estimation step by least squares method. The proposed algorithm was evaluated by a simulation with a measured data from CIPIC (Center for Image Processing and Integrated Computing) HRTF database, and the simulation results show that the proposed algorithm reduces mean-squared error by 1.5 dB ~ 4 dB than the conventional algorithms.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.6 s.44
• /
• pp.269-277
• /
• 2006

We propose a new expert system for recovering the broken fragments of relics into an original form using computer graphics and image processing. This paper presents a system with an application to tombstones objects of flat plane with letters carved in for assembling the fragments by placing their respective fragments in the right position. The matching process contains three sub-processes: aligning the front and letters of an object, identifying the matching directions, and determining the detailed matching positions. We apply least squares fitting, vector inner product, and geometric and RGB errors to the matching process. It turned out that 2-D translations via fragments-alignment enable us to save the computational load significantly. Based on experimental results from the damaged cultural fragments, the performance of the proposed method is illustrated.

• Journal of the Economic Geographical Society of Korea
• /
• v.10 no.4
• /
• pp.414-426
• /
• 2007

The purpose of this study is to analyze whether the spatial externalities of agglomeration economy of manufacturing has been changed or not, from 1991 to 2005. To find this spatial SUR model was used to consider covariance of each equation and it was estimated by 3SLS. This paper found that spatial externalities has been decreased and there is economy of scale in same sector. As a result, we can conclude that there is industrial specialization in region whereas interregional linkage are weaken. This finding supports the desirability of creating supra-regional agencies promoting interregional linkage because supra-regional agencies can internalize spatial externality of agglomeration economy.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06a
• /
• pp.76-78
• /
• 2006

현재 심혈관 질환은 암 다음으로 높은 사망 원인으로 기록되고 있어 심혈관 질환에 대한 초기 진단은 질환의 치료에 매우 중요한 문제로 대두되고 있다. 본 논문에서는 SVM을 이용하여 심혈관질환 환자의 질환 단계를 예측하였다. 일반적으로 이진분류에 사용되는 SVM을 이용하여 정상 및 질환 $1{\sim}3$기의 총 4가지 분류가 필요한 다분류 분류문제를 처리하기 위해서 논문에서는 독립적 학습된 단일 SVM 분류기들을 결합하여 분류를 수행하는 SVM 앙상블 방법을 사용하였다. 단일 분류기의 결합은 Majority voting, 최소자승에러기반 가중치 부여, 2단계층 결합 등의 방법으로 수행하여 심혈관 질환 분류에 적합한 앙상블의 구성을 시도하였다. 실험 데이터는 (주)제노프라의 압타머 칩 데이터를 사용하였다. 서로 다른 데이터를 이용하여 학습된 이종의 SVM들을 결합한 결과 질환단계 예측에 있어서 단일 SVM을 이용하여 질환 단계를 예측하는 경우 보다 향상된 질환단계 예측 성능을 관찰할 수 있었으며, 심혈관 질환의 예측에 대해서는 단일 SVM 분류기의 2단 계층 결합법이 가장 좋은 성능을 보임을 확인하였다.

• Environmental and Resource Economics Review
• /
• v.6 no.2
• /
• pp.259-274
• /
• 1997

본 논문은 쓰레기종량제 하에서 가격유인책 및 다른 사회 경제적 변수가 가정의 쓰레기발생과 재활용품배출에 미치는 효과를 분석하고 있다. 개별 가정설문자료를 이용하여 가정의 쓰레기발생과 재활용품배출에 대한 구조적 방정식을 3단계 최소자승법으로 추정하였다. 추정결과 가계소득과 가족 수는 쓰레기발생량에 정의 관계로 영향을 미치며, 재활용품배출량은 가정주부의 시간가치와는 역의 관계를 가지고 교육수준과는 정의 관계를 보이고 있다. 쓰레기봉투가격의 인상은 쓰레기발생량에는 영향을 주지 않고 재활용품배출량의 증가를 통해 쓰레기수거서비스에 대한 수요를 감소시키는 것으로 나타난다. 가정에서 분리수거를 통해 재활용할 수 있는 잠재적 재활용가능량은 정부의 재활용품목의 지정 및 재활용기술에 의해 제한되기 때문에 쓰레기가격의 대폭적 인상을 통해 쓰레기수거서비스의 감소를 유도하는 것은 한계가 있는 것으로 보인다.

• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.145-152
• /
• 2021

In this study, we have examined the effect of open innovation of SMEs on innovation performance of firms. Most studies do not consider internal generation in relation to open innovation and innovation performance.We conducted empirical studies to overcome this problem. The research was carried out by collecting data collected from 512 SMEs and the 3SLS method was used to minimize the internal generation. As a result, open innovation investment and use of external ideas among SMEs' open innovation have positively influenced project success and technical performance. But, the introduction of technology and cooperation with the research organization did not affect the innovation performance, which is presumed to be due to the fact that the open innovation of SMEs is limited to relatively inexpensive activities. Therefore, in order to promote open innovation of SMEs, it is necessary to provide support for relatively high-cost activities and to improve the innovation performance of enterprises. In order to reduce the difficulties of open innovation activities, domestic universities and research institutes should support the construction of enterprise networks and actively support the utilization of technology to expand innovation performance.

• Journal of Institute of Control, Robotics and Systems
• /
• v.20 no.4
• /
• pp.414-421
• /
• 2014

This paper presents a method of ballistic missile launch point estimation using phase division least squares. The proposed algorithm employs smoothing to enhance estimation accuracy and generates functions of time for total velocity, flight path angle and heading angle, allowing extrapolation to estimate the launch point. Performance of the proposed algorithm is tested in conjunction with the extended Kalman filter and the Kalman filter.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.565-568
• /
• 2018

본 연구에서는 Lasso Regression을 기반으로 하여 지역 경제 성장과 비만율을 예측한다. 연구는 3단계로 나누어 진행된다. 우선 지역성장을 대변할 수 있는 가상의 GDP 수치를 구한다. 그 다음 가상의 GDP 수치와 비만율 데이터를 이용하여 학습모델을 만든다. 마지막으로 이전의 데이터를 이용하여 앞으로의 성장을 예측하고 학습모델에 적용하여 비만율을 예측한다. 본 연구의 데이터는 학습데이터와 실험데이터를 구성된다. 학습데이터로는 국내의 8도 중 하나인 강원도의 데이터를 이용하며 실험데이터로는 강릉과 원주의 데이터를 이용한다. 평가 비교 대상으로는 과거의 흐름을 반영하는 최소자승법 예측기법을 선정하여 비교한다. 연구 결과 강릉의 경우 비교 데이터와의 오차율 평균은 1.22%로 큰 차이가 없음을 알 수 있다. 따라서 본 연구에서 제안하는 방법이 과거의 흐름을 기반으로 작성됨을 알 수 있다. 하지만 단순히 과거의 흐름만을 통해 예측하는 것은 여러 요소가 복합적으로 작용하는 비만율 예측에 알맞지 않기 때문에 본 연구 방법이 유의미하다고 여겨진다.

• International Area Studies Review
• /
• v.12 no.3
• /
• pp.25-47
• /
• 2008

This paper tries to explore the impact of economic globalization on social capital accumulation. To investigate direct effect, we build a model and derive a proposition which can explain the relative decline in social capital brought about by market expansion. Besides direct effect, we also explore channel effect through democracy, government size, education attainment, and inequality. We estimate direct and channel effect of globalization using cross section of 65 countries data time period 1980-1999 using three-stage least squares(3SLS). Results are in line with predictions and clearly support that globalization significantly and negatively affects social capital accumulation. Channel effect also shows that globalization has a negative effect through aggravating income inequality while it has a positive effect through higher education attainment, higher level of democracy and larger government spending. Such a net negative channel effect reinforces our prediction. As a robustness check we estimate other sets of data and the result strongly supports our theory.