• Communications for Statistical Applications and Methods
• /
• 제15권2호
• /
• pp.283-292
• /
• 2008

커널밀도함수의 추정을 이용한 분류 문제에서 평활모수(smoothing parameter, bandwidth)의 선택은 핵심적으로 중요한 역할을 한다. 본 논문에서는 분류에서 베이즈 리스크를 최적화하기 위한 평활모수의 선택이 각 개별 확률밀도함수를 추정하기 위한 최적의 평활모수와 어떤 관계가 있는지 살펴보았다. 실제 상황에서 사용할 수 있는 평활모수의 선택 방법으로 붓스트랩(bootstrap)과 교차확인법(cross-validation)을 이용하는 것을 비교한 결과, 붓스트랩 방법은 Hall과 Kang (2005)에서 밝혀진 이론적인 성질에 부합하는 반면 교차확인법은 그렇지 못함을 확인하였다. 또한, 각 방법으로 정한 평활모수를 사용하여 오분류율을 조사해 본 결과에서도 붓스트랩 방법이 우월함을 알 수 있었다.

• KSBB Journal
• /
• 제17권4호
• /
• pp.326-332
• /
• 2002

크로마토그래피 공정 성능 검증의 한 항목인 담체의 수명 검증을 위해 자동화된 미니 크로마토그래피 시스템(MiniValChrom)을 개발하였다. 이 시스템은 크로마토그래피공정의 자동화된 반복수행, 크로마토그래피 각 step의 작동순서 및 공정조건의 자유로운 구성, on-line 실시간 모니터링 및 제어, 여러 담체 수명 검증법의 method file 저장기능 등을 갖추었다. MiniValChrom을 사용하여 BSA와 Cibacron Blue 3G-A를 각각 모델 단백질과 담체로 담체 수명 검증실험을 사례연구로 수행하였다. 담체 수명의 감소는 크로마토그래피 공정을 반복수행하면서 5 cycle 마다 변화하는 HETP값을 측정하여 HETP 값이 1 cm 이상 될 때까지 반복 수행함으로써 담체 수명을 결정할 수 있었다. 본 연구를 통해 개발된 MiniValChrom은 다른 검증 항목이나 multi-product 생산공장 내 크로마토그래피 공정 검증에도 유용하게 쓰일 수 있을 것으로 기대된다.

• 대한가정학회:학술대회논문집
• /
• 대한가정학회 1997년도 Program & Abstracts * The 7th Korea-Japan Home Economics Joint Syposium
• /
• pp.269-269
• /
• 1997

• 아동학회지
• /
• 제37권5호
• /
• pp.129-144
• /
• 2016

Objective: This study's purpose was to develop a teaching competency scale for character education for early childhood teachers and to examine its reliability. Methods: The study procedure was as follows: First, scale items were generated by reviewing the literature as a deductive approach and focused interviews with 8 field experts in an inductive approach. Second, face validation was assessed by 7 academic experts, and statistical validation processes were done twice. The data used in the final statistical analyses included 206 questionnaires for early childhood teachers. Statistical validation included item response analysis, item discrimination analysis, exploratory factor analysis, confirmatory factor analysis, and reliability analysis. Results: The 3 factors identified were as follows: Communicate teaching competence, practical teaching competence, and instrumental teaching competence. The validity and reliability of the Teaching Competency Scale for Character Education in Early Childhood Teachers were acceptable. Conclusion: This new scale will be a useful tool to better support character education for early childhood teachers and contribute to providing a basis for developing more sophisticated tools.

• Wind and Structures
• /
• 제11권1호
• /
• pp.51-70
• /
• 2008

CFD is applied to evaluate pedestrian wind comfort at outdoor platforms in a high-rise apartment building. Model validation is focused on generic building sub-configurations that are obtained by decomposition of the actual complex building geometry. The comfort study is performed during the design stage, which allows structural design changes to be made for wind comfort improvement. Preliminary simulations are performed to determine the effect of different design modifications. A full wind comfort assessment study is conducted for the final design. Structural remedial measures for this building, aimed at reducing pressure short-circuiting, appear to be successful in bringing the discomfort probability estimates down to acceptable levels. Finally, the importance of one of the main sources of uncertainty in this type of wind comfort studies is illustrated. It is shown that the uncertainty about the terrain roughness classification can strongly influence the outcome of wind comfort studies and can lead to wrong decisions. This problem is present to the same extent in both wind tunnel and CFD wind comfort studies when applying the same particular procedure for terrain relation contributions as used in this paper.

• 한국컴퓨터정보학회논문지
• /
• 제27권11호
• /
• pp.13-18
• /
• 2022

본 논문에서는 학습에 참여하는 각 디바이스의 모델들로부터 성능검증에 따라 가중치를 두어 글로벌 모델을 업데이트하는 VW-FedAVG(Validation based Weighted FedAVG)를 두 가지 방식으로 제안 한다. 첫 번째 방식은 서버 검증(Server side Validation) 구조로 글로벌 모델을 업데이트 하기 전에 각 로컬 클라이언트 모델을 하나의 전체 검증 데이터셋을 통해 검증하도록 설계 했다. 두 번째는 클라이언트 검증(Client side Validation) 구조로 검증 데이터셋을 각 클라이언트에 고르게 분배하여 검증을 한 후 글로벌 모델을 업데이트 하는 방식으로 설계 했다. 전체 실험에 적용한 데이터셋은 MNIST, CIFAR-10으로 이미지 분류에 대해 IID, Non-IID 분포에서 기존 연구 대비 더 높은 정확도를 얻을 수 있었다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 2010년도 정기 학술대회
• /
• pp.379-382
• /
• 2010

This study proposes a way to validate BIM data quality in BIM applications. Solibri model checker is adopted as a module development platform, which is based on Java programming language. The platform makes application developers implement BIM model checker for their own purpose. This study has developed a BIM validation module for circulation analysis of building design. The validation module enables end-users to automatically detect data corrupted or not defined. In case studies, the module found that an IFC file generated from a BIM software has wrong relation information between a space and boundary elements. A building model should satisfy modeling requirements and then domain users can get analysis results. The BIM data validation module needs to be developed in each BIM application domain.

• Nuclear Engineering and Technology
• /
• 제54권4호
• /
• pp.1230-1244
• /
• 2022

This paper proposes an algorithm for signal validation using unsupervised methods in emergency situations at nuclear power plants (NPPs) when signals are rapidly changing. The algorithm aims to determine the stuck failures of signals in real time based on a variational auto-encoder (VAE), which employs unsupervised learning, and long short-term memory (LSTM). The application of unsupervised learning enables the algorithm to detect a wide range of stuck failures, even those that are not trained. First, this paper discusses the potential failure modes of signals in NPPs and reviews previous studies conducted on signal validation. Then, an algorithm for detecting signal failures is proposed by applying LSTM and VAE. To overcome the typical problems of unsupervised learning processes, such as trainability and performance issues, several optimizations are carried out to select the inputs, determine the hyper-parameters of the network, and establish the thresholds to identify signal failures. Finally, the proposed algorithm is validated and demonstrated using a compact nuclear simulator.

• 대한약침학회지
• /
• 제12권4호
• /
• pp.33-50
• /
• 2009

Objectives : This study was conducted to confirm validation and stability of concentration analysis method of pure melittin (Sweet Bee Venom-Sweet BV) extracted from the bee venom by utilizing protein isolation method of gel filtration. Methods : All experiments were conducted at Biotoxtech, a non-clinical studies authorized institution, under the regulations of Good Laboratory Practice (GLP). Standard solutions of melittin (SIGMA, USA) and test substances were dispensed and were analyzed with HPLC for Sweet BV to secure the validation of analysis. Results : 1. Measurement of system suitability of Sweet BV satisfied criterion of below 3%. 2. Confirming Linearity of Sweet BV in 10-200${\mu}g/m\ell$ solution yielded correlation coefficient (r) of 0.995 and accuracy of 85-115% which satisfy criterion. 3. Measurement of Specificity of Sweet BV didn't yield any substance affecting the peak of test substances, but detected at 21.22min verified as the test substance. 4. Confirming Intra-day of Sweet BV, accuracy and precision of 0.1, 100${\mu}g/m\ell$ were 105.70, 95.81 and 0.66, 0.73, respectively, satisfying both criteria of accuracy (85-115%) and precision (within 10%). 5. To measure Stability in autosampler, all samples used in Intra-day reproducibility sat in the autosampler for five hours and were re-analyzed. Both variability and precision satisfied the criteria. 6. Homogeneity of Sweet BV (0.1, 100${\mu}g/m\ell$) at upper, middle, and lower layers all satisfied the accuracy and precision criteria. 7. Stability of Sweet BV (0.1, 100${\mu}g/m\ell$) at room temperature for four hours and refrigerated for 7 days all satisfied the criterion. 8. For the measurement of Quality control, QC samples measured on the first and eighth day all satisfied accuracy and precision criteria. Conclusion : Above experiment data satisfies validation and stability of concentration analysis method of Sweet BV.

• Ocean and Polar Research
• /
• 제30권1호
• /
• pp.47-58
• /
• 2008

Several validation studies have been made for QuikSCAT(QSCAT) wind data around the world, mainly in the offshore. However, until now, there were no validation studies for QSCAT wind with resolution of 12.5 km ('QSCAT 12.5 km wind') in the vicinity of Korean Peninsula. To validate 'QSCAT 12.5 km wind' and to investigate its characteristics around Korean Peninsula, the wind data from Ieodo Ocean Research Station, KMA buoys, and KORDI Realtime Observation Stations have been compared. Validation results showed that 'QSCAT 12.5 km wind' RMSE of wind direction and speed were $25.85^{\circ}$ and 1.83 m/s, respectively, at Ieodo Station. The mean wind speed correlation coefficient of KMA buoys and KORDI Realtime Observation Station were 0.78 and 0.61, and the mean wind speed RMSE were 2.2 m/s and 3.2 m/s, respectively. This seems to be mainly because of the distance between QSCAT and in-situ observation stations. The RMSE of wind direction were bigger than $40^{\circ}$ at all in-situ observation stations located near the shore, within 20 km from coastlines. Geophysical features where in-situ observation stations are located seem to affect wind validation scores.