• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.33 no.4
• /
• pp.409-416
• /
• 2009

Standard procedures for a fracture toughness testing require very severe restrictions for the specimen geometry to eliminate a size effect on the measured properties. Therefore, the used standard fracture toughness data results in the integrity assessment being irrationally conservative. However, a realistic fracture in general structures, such as in nuclear power plants, may develop under the low constraint condition of a large scale yielding with a shallow surface crack. In this paper, cleavage fracture toughness tests have been made on side-grooved PCVN (precracked charpy V-notch) type specimens (10 by 10 by 55 mm) with various crack depths. The constraint effects on the crack depth ratios were evaluated quantitatively by the developed scaling method using the 3-D finite element method. After the fracture toughness correction from scaling model, the statistical size effects were also corrected according to the standard ASTM E 1921 procedure. The results were evaluated through a comparison with the $T_0$ of the standard CT specimen. The corrected $T_0$ for all of the PCVN specimens showed a good agreement to within $5.4^{\circ}C$ regardless of the crack depth, while the averaged PCVN $T_0$ was $13.4^{\circ}C$ higher than the real CT test results.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.123-129
• /
• 2001

지금까지 한국어 품사 부착을 위해 다양한 모델이 제안되었고 95% 이상의 높은 정확도를 보여주고 있다. 그러나 4-5%의 오류는 실제 응용 분야에서 많은 문제를 야기시킬 수 있다. 이러한 오류를 최소화하기 위해서는 오류를 분석하고 이를 수정할 수 있는 규칙들을 학습하여 재사용하는 방범이 효과적이다. 오류 수정 규칙을 학습하기 위한 기존의 방법들은 수동학습 방법과 자동 학습 방법으로 나눌 수 있다 수동 학습 방법은 많은 비용이 요구되는 단점이 있다. 자동 학습 방법의 경우 모두 변형규칙 기반 접근 방법을 사용하였는데 어휘 정보를 고려할 경우 탐색 공간과 규칙 적용 시간이 매우 크다는 단점이 있다. 따라서 본 논문에서는 초기 모델에 대한 오류 수정 규칙을 효율적으로 학습하기 위한 새로운 방법으로 결정 트리 학습 방법을 확장한 통계적 결정 그래프 학습 방법을 제안한다. 제안된 방법으로 두 가지 실험을 수행하였다. 초기 모델의 정확도가 높고 말뭉치의 크기가 작은 첫 번째 실험의 경우 초기 모델의 정확도 95.48%를 97.37%까지 향상시킬 수 있었다. 초기 모델의 정확도가 낮고 말뭉치 크기가 큰 두 번째 실험의 경우 초기 모델의 정확도 87.22%를 95.59%로 향상시켰다. 또한 실험을 통해 결정 트리 학습 방법에 비해 통계적 결정 그래프 학습 방법이 더욱 효과적임을 알 수 있었다.

• Nuclear Engineering and Technology
• /
• v.32 no.4
• /
• pp.361-371
• /
• 2000

The 123Te(p,n)123I, 124Te(p,n)124I and 124Te(p,2n)123I reactions, among the many reaction channels opened, are the major reactions under consideration from a diagnostic purpose because reaction residuals as the gamma emitters are used for most radiophamaceutical applications involving radioiodine. Based on the available experimental data, the absorption cross sections and elastic scattering angular distributions of the proton-induced nuclear reaction on Te isotopes below 60 MeV are calculated using the optical model code APMNK. The transmission coefficients of neutron, proton, deuteron, trition and alpha particles are calculated by CUNF code and are fed into the GNASH code. By adjusting level density parameters and the pair correction values of some reaction channels, as well as the composite nucleus state density constants of the pre-equilibrium model, the production cross sections and energy-angle correlated spectra of the secondary light particles, as well as production cross sections and energy distributions of heavy recoils and gamma rays are calculated by the statistical plus pre-equilibrium model code GNASH. The calculated results are analysed and compared with the experimental data taken from the EXFOR. The optimized global optical model parameters give overall agreement with the experimental data over both the entire energy range and all tellurium isotopes.

• The Korean Journal of Applied Statistics
• /
• v.17 no.2
• /
• pp.239-252
• /
• 2004

It is known that the domestic and the U.S. stock prices tend to move together as those markets are closely interrelated. In this study, cointegration and causal relationships among the four stock price indexes of KOSPI, KOSDAQ, DOWJONES and NASDAQ are carefully investigated for the period of declining stock prices in the long run. When all indexes move in a similar fashion, cointegration does not exist and the causal linkages between the domestic and the U.S. stock prices appear relatively complex. On the other hand, when the domestic and the V.S. stock prices move in a different manner, cointegration exists and the causal relationships appear relatively simple. NASDAQ is apparently found to lead the domestic stock market in both periods, which is consistent with the actual market situation when the If industry is under recession.

• Analytical Science and Technology
• /
• v.25 no.4
• /
• pp.214-222
• /
• 2012

ASTM (American Society for Testing and Materials) D6751-10 suggests analytical methods as well as specifications for biodiesel quality. However, it is expensive and time-consuming to follow the ASTM testing methods to analyze biodiesel and various impurities. This paper develops a quantitative analysis system for biodiesel and impurities based on Infrared spectroscopy and a multivariate statistical method, PLS (partial least squares). In addition, four different pre-processing techniques were compared for spectrum correction and noise reduction. Savitzky-Golay pre-processing showed the best performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.5 no.3
• /
• pp.560-574
• /
• 2011

In this paper, we propose a hybrid method of Machine Learning (ML) algorithm and a rule-based algorithm to implement a lightweight Named Entity (NE) extraction system for Korean SMS text. NE extraction from Korean SMS text is a challenging theme due to the resource limitation on a mobile phone, corruptions in input text, need for extension to include personal information stored in a mobile phone, and sparsity of training data. The proposed hybrid method retaining the advantages of statistical ML and rule-based algorithms provides fully-automated procedures for the combination of ML approaches and their correction rules using a threshold-based soft decision function. The proposed method is applied to Korean SMS texts to extract person's names as well as location names which are key information in personal appointment management system. Our proposed system achieved 80.53% in F-measure in this domain, superior to those of the conventional ML approaches.

• Structural Engineering and Mechanics
• /
• v.75 no.5
• /
• pp.633-642
• /
• 2020

Artificial neural networks (ANNs) are known as intelligent methods for modeling the behavior of physical phenomena because of it is a soft computing technique and takes data samples rather than entire data sets to arrive at solutions, which saves both time and money. ANN is successfully used in the civil engineering applications which are suitable examining the complicated relations between variables. Functionally graded materials (FGMs) are advanced composites that successfully used in various engineering design. The FGMs are nonhomogeneous materials and made of two different type of materials. In the present study, the bending analysis of functionally graded material (FGM) beams presents on theoretical based on combination of mixed-finite element method, Gâteaux differential and Timoshenko beam theory. The main idea in this study is to build a model using ANN with four parameters that are: Young's modulus ratio (E_t/E_b), a shear correction factor (k_s), power-law exponent (n) and length to thickness ratio (L/h). The output data is the maximum displacement (w). In the experiments: 252 different data are used. The proposed ANN model is evaluated by the correlation of the coefficient (R), MAE and MSE statistical methods. The ANN model is very good and the maximum displacement can be predicted in ANN without attempting any experiments.

• Maxillofacial Plastic and Reconstructive Surgery
• /
• v.22 no.3
• /
• pp.310-318
• /
• 2000

The purpose of this study was to assess the soft tissue changes of upper lip & nose using 38 patients who treated with Le Fort I osteotomy for the correction of dentofacial deformities. Patients were devided into three groups. One was advancement group of maxilla(Group I, N=14), another was impaction group of maxilla(Group II, N=12) and the other was combination group(advancement & impaction)(Group III, N=12). Preop. and 1 month postop. (T1), preop. and 6 months postop.(T2) were analyzed and compared. The results obtained were as follows : 1. The upper lip thickness(UL-VP) moved anteriorly approximately 62% of the horizontal maxillary change and this was significant in the advancement group(Group I) 2. The upper lip length(Stm-Sn) and the lower border of upper lip(Stm) moved superiorly 25%, 40% of the maxillary impaction group(Group II) (P<0.05) 3. There was significancy in the upper lip thicness(UL-VP) approximately 56% of the combination group(Group III) (P<0.05) 4. The nasolabial angle decreased in all groups, but there were no significancy.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.7 no.2
• /
• pp.1-6
• /
• 1989

The adjustment of levelling net is being done to the order of nets independently by using the least square method. For the small size net, it has difficulties in verification and statistical analysis of the net since the degree of freedom is low At the same time, it is also difficult to evaluate the error of lower order net correctly. The aim of this study is to analyse the properties of combined adjustment method compared with the independent adjustment method by using the data which have been measured during 1967-1987. Another aim is to analyse the influences of normal orthometric correction and changes of datum. Finally, Korean leveling net has been evaluated by applying real redundancy and variance component estimation.

• The Korean Journal of Applied Statistics
• /
• v.32 no.3
• /
• pp.405-421
• /
• 2019

Predicting how much energy to use is an important issue in society. However, it is more difficult to capture the usage characteristics of residential buildings than other buildings. This paper provides time series analysis methods for electricity consumption in a residential building. Temperature is closely related to electricity demand. An error correction model, which is a method of adjusting the error with time, is applied when a cointegration relation is established between variables. Therefore, we analyze data via ECMs with consideration of the temperature effect.