Earticle

The estimation of missing sensor values is an important problem in sensor network applications, but the existing approaches have some limitations, such as the limitations of application scope and estimation accuracy. Therefore, in this paper, we propose a new estimation model based on a spatial-temporal correlation analysis (STCAM). STCAM can make full use of spatial and temporal correlations and can recognize whether the sensor parameters have a spatial correlation or a temporal correlation, and whether the missing sensor data are continuous. According to the recognition results, STCAM can choose one of the most suitable algorithms from among linear interpolation algorithm of temporal correlation analysis (TCA-LI), multiple regression algorithm of temporal correlation analysis (TCA-MR), spatial correlation analysis (SCA), spatial-temporal correlation analysis (STCA) to estimate the missing sensor data. STCAM was evaluated over Intel lab dataset and a traffic dataset, and the simulation experiment results show that STCAM has good estimation accuracy.

Recent advances in deep learning have facilitated widespread deployment of posture recognition system. However, the scarcity of accessible human posture datasets presents a significant challenge to achieving high model precision and robustness. To address this challenge, we introduce a federated learning (FL) framework that enables training of posture recognition models across diverse and privacy-sensitive datasets. Our approach integrates FL strategies with fundamental deep learning architectures for human pose estimation and action recognition. To identify the optimal base model, we evaluate five widely adopted architectures. Using a posture recognition dataset, our systematically compare conventional single-machine training models with FL-enhanced models across key metrics: precision, recall, F1-score, and accuracy. Results demonstrate that our FeLM achieves superior generalization while preserving data privacy, highlighting the efficacy of FL in posture recognition tasks. This underscores the benefit of FL in improving the generalization of a model without compromising data privacy.

This study examined those problems noticed under the Asymptotically Generalized Least Squares estimator in evaluating a structural model of physical health. The problems were highly correlated parameter estimates and high standard errors of some parameter estimates. Separate analyses of the endogenous part of the model and of the metric of a latent factor revealed a highly skewed and kurtotic measurement indicator as the focal point of the manifested problems. Since the sample sizes are far below that needed to produce adequate AGLS estimates in the given modeling conditions, the adequacy of the Maximum Likelihood estimator is further examined with the robust statistics and the bootstrap method. These methods demonstrated that the ML methods were unbiased and statistical decisions based upon the ML standard errors remained almost the same. Suggestions are made for future studies adopting structural equation modeling technique in terms of selecting of a reference indicator and adopting those statistics corrected for nonormality.

This study addresses the accurate prediction of weld-bead geometry, which is critical for improving weld quality, reducing potential defects, and optimizing operational efficiency. To overcome the challenges of traditional inverse models, we present a diffusion-based framework that incorporates temporal guidance through sequential frame data. Our model integrates a guidance encoder, latent transform with flow-based modeling, and a diffusion decoder to produce temporally consistent and structurally faithful 3D point clouds. The proposed model utilizes condition information to restore weld geometry data more accurately and its guided learning structure effectively supports overall shape preservation, spatial consistency, and time series-based continuous shape generation. Experiments showed that it outperforms standard diffusion methods using the PSNR, Chamfer distance, Jensen-Shannon divergence, and Hungarian error metrics. Qualitative results showed that the model maintains a balanced depiction of the global geometry without over-emphasizing areas with high point density. These results highlight potential applications in VR-based welding training systems and real-time adaptive control of automated welding processes.

Purpose: The purpose of this study was to investigate Korean preschoolers' obesity-related factors through an ecological approach and to identify Korean preschoolers' obesity-related factors and the different effects of ecological variables on body mass index and its quantiles through an ecological approach. Methods: The study design was cross-sectional. Through convenience sampling, 241 cases were collected from three kindergartens and seven nurseries in the Seoul metropolitan area and Kyunggi Province in April 2013 using self-administered questionnaires from preschoolers' mothers and homeroom teachers. Results: Results of ordinary least square regression analysis show that mother's sedentary behavior (p < .001), sedentary behavior parenting (p = .039), healthy eating parenting (p = .027), physical activity related social capital (p = .029) were significant factors of preschoolers' body mass index. While in the 5% body mass index distribution group, gender (p = .031), preference for physical activity (p = .015), mother's sedentary behavior parenting (p = .032), healthy eating parenting (p = .005), and teacher's sedentary behavior (p = .037) showed significant influences. In the 25% group, the effects of gender and preference for physical activity were no longer significant. In the 75% and 95% group, only mother's sedentary behavior showed a statistically significant influence (p < .001, p = .012 respectively). Conclusion: Efforts to lower the obesity rate of preschoolers should focus on their environment, especially on the sedentary behavior of mothers, as mothers are the main nurturers of this age group.

The objectives of this study were to quantify the combination values of the principal components and factors calculated using body measurements of Hanwoo(Korean Cattle) and estimate their heritabilities. The technique of multivariate analysis was used to reduce a large number of variables to a smaller number of new variables and characterize cattle according to body shape. The analyses were performed using 1,979 cattle at 12 months of age and 936 cattle at 24 months of age. The data for the analyses was obtained from progeny tests performed on Korean Cattle for 6 years from 2003 to 2008. The phenotypic correlations among these traits were estimated to range from 0.32 to 0.90 at 12 months of age and from 0.21 to 0.82 at 24 months of age. The first principal components(PC1s) indicated a weighed average of overall body measurements, accounting for 99.91% of the total variation for both periods of test. The two first PCs had positive coefficients for all body measurements. The major sources of PC, such as chest girth(CG), body length(BL), rump height(RH), and wither height(WH) were similar for both test periods. The heritabilities for PC1, the first factor score(FS1), and the second factor score(FS2) were estimated by multivariate REML method. The estimated heritabilities for PC1, FS1, and FS2 were 0.33, 0.38, and 0.40, respectively, at 12 months of age and 0.26, 0.76, and 0.58 at 24 months of age. Further studies are needed to determine whether the heritabilities of FS1 and FS2 at 24 months of age were overestimated.

본 연구는 임의회귀모형을 이용하여 한우 거세우 체중에 대해서 유전모수 추정을 하고 이것을 단형질 개체모형의 결과와 비교해 보고자 실시하였다. 분석에 이용한 자료는 총 1,372두의 한우 거세우의 체중 자료로, 농협중앙회 가축개량사업소에서 실시한 한우 후대검정우의 기록이다. 이차의 임의회귀 모형에 적용한 결과 유전력이 총 800일령까지의 검정기간에 대해 0.17~0.30의 범위로 나타났다. 개체모형을 통해 얻은 유전력은 0.24~0.36의 범위로 나타났다. 측정일간의 영구환경효과의 상관은 검정일령이 늘어남에 따라 함께 증가하는 경향을 보였다. 반면, 측정일간의 유전상관의 경우 검정초기에는 0.30 정도의 약한 음의 상관을 보이지만, 검정이 이뤄짐에 따라 상관이 점차 증가하여 검정이 종료될 무렵이면 거의 고정되는 것으로 나타났다. 임의회귀모형과 개체모형의 결과를 비교해보면 두가지 모형 모두 비슷한 경향을 보여 큰 차이를 보이지 않았다. 따라서, 임의회귀모형을 한우에 대한 국가유전능력평가에 사용하는 것이 가능할 것으로 사료된다.

The study aimed to estimate genetic parameters of body weights in Hanwoo steers using random regression model and compare it with single trait animal model. A total of 1,372 Hanwoo steers that belonged to progeny testing program of the Hanwoo Genetic Improvement conducted at the Livestock Improvement Main Center of the National Agricultural Cooperative Federation (LIMC-NACF) in Rep. of Korea were used. Results of the random regression model fitting quadratic function revealed heritability values from 0.17 to 0.30 for the whole testing days up to 800 days. The results of the animal model showed estimated heritability values ranged from 0.24 to 0.36. Estimates of permanent environmental correlations tended to increase with increasing test in days. Unlike in the direct genetic correlation that at early stage the estimate was slightly negative it was 0.30 then increased to approach unity at later stage of test. Comparing the results between random regression model and the animal model showed not much differences and both followed similar pattern and therefore the use of random regression model for the national genetic evaluation of Hanwoo could be implemented.

This study aimed to develop a hybrid stem taper and volume estimation model using multi-platform LiDAR data for Pinus koraiensis (PK) and Larix kaempferi (LK) in a forest located in Gangwon Province, Republic of Korea. The research employed Terrestrial Laser Scanning (TLS) to capture detailed point cloud data of tree stems and Airborne Laser Scanning (ALS) for precise height measurements. By integrating the stem profiles derived from TLS with Kozak’s stem taper model, a hybrid estimation model was constructed to improve the accuracy of calculating individual tree taper curves and stem volumes. The proposed model was tested against traditional volume estimation methods, specifically the Korea Forest Service's standard volume table, to assess its accuracy. The standard volume table method exhibited root mean square error (RMSE) values of 0.12 m³ for PK and 0.13 m³ for LK. In contrast, the hybrid model showed significantly lower RMSE values of 0.07 m³ for PK and 0.05 m³ for LK, representing an accuracy improvement of approximately 42% for PK and 62% for LK. Additionally, the study estimated log production based on individual tree profiles generated from the hybrid model, accounting for factors such as stem length, diameter, and curvature. The increased accuracy of the hybrid model highlights its potential for providing more precise stem volume estimations and improving the reliability of forest inventories. The results suggest that this hybrid approach, leveraging the strengths of both TLS and ALS, offers a more efficient and accurate alternative to traditional volume estimation methods. The precise taper curves and volume estimates derived from this method can significantly contribute to sustainable forest management practices, particularly in assessing timber production and monitoring forest growth. By eliminating the need for destructive sampling, this method provides a non-invasive solution for estimating tree volumes. In conclusion, the multi-platform LiDAR-based hybrid model offers a promising solution for more accurate stem volume estimation, supporting forest inventory development and enhancing decision-making in forest management and timber production.

본 연구에서는 공공기관에서 제공하고 있는 정량적 자료를 이용하여 자연재해로 인한 피해액을 추 정하기 위한 회귀분석 모형을 제안한다. 이를 위해, 공공 부문에서 225개 기초지방자치단체에 대해 2001년부터 2014년까지의 자연재해 세부지표 값과 피해액 자료를 수집하여 회귀분석을 수행한다. 본 연구에서는 회귀모형이 사용하는 독립변수의 개수를 줄이기 위해 변수선정 방법론을 활용한다. 이를 통해, 기존 자연재해 위험지표에서 사용하던 32개 세부지표 중 주요 변수만을 선정하여 14개 독립변수로 구성된 자연재해 피해액 추정 모형을 정의하였다. 제안된 모형의 유효성을 검증하기 위해, 모형을 이용하여 추정된 피해액 예측치, 자연재해 위험지표 평가결과, 국민안전처 지역안전도 평가결과, 실제 피해액을 비교⋅분석하였다. 분석 결과, 본 연구에서 도출된 피해액 예측치가 기존 자연재해 위험지표와 지역안전도에 비해 자연재해 피해액과 더 높은 상관관계를 갖는 것을 확인할 수 있었다. 본 연구에서 제안하고 있는 회귀분석 기반 자연재해 피해액 추정 모형을 활용하면 향후 자연재해 발생 양상의 변화로 인한 자연재해 피해액의 변화를 미리 예측할 수 있으며, 제안된 회귀 분석 모형의 회귀계수를 가중치로 활용하면 기존 자연재해 위험지표의 성능 및 정확도를 향상시킬 수 있을 것이다.

In this study, we propose a regression model for estimating damage amounts of natural disasters based on the quantitative data provided by public institutions. Based on quantitative data for 225 primary local governments during the years of 2001 to 2014, we define a regression model to have 14 independent variables by applying variable selection methods to the variables from 32 sub-indexes used in Natural Disaster Risk Index(NDRI). For verifying the proposed regression model, we compare the estimates from the model with NDRI, Regional Safety Grades(RSG), and actual amounts of damages from natural disasters. From the analysis results, we can find the regression model give better estimates for the amounts of damages than NDRI and RSG. If we utilize the proposed model, we can estimate the future changes in the amounts of damages using the changes in natural disasters and improve the performance of NDRI using the standard regression coefficients from the regression analysis.

Background: After the Fukushima Daiichi Nuclear Power Plant (FDNPP) accident, biological alterations in the natural biota, including morphological changes of fir trees in forests surrounding the power plant, have been reported. Focusing on the terminal buds involved in the morphological formation of fir trees, this study developed a method for estimating the absorbed radiation dose rate using radionuclide distribution measurements from tree organs. Materials and Methods: A phantom composed of three-dimensional (3D) tree organs was constructed for the three upper whorls of the fir tree. A terminal bud was evaluated using Monte Carlo simulations for the absorbed dose rate of radionuclides in the tree organs of the whorls. Evaluation of the absorbed dose targeted 131I, 134Cs, and 137Cs, the main radionuclides subsequent to the FDNPP accident. The dose contribution from each tree organ was calculated separately using dose coefficients (DC), which express the ratio between the average activity concentration of a radionuclide in each tree organ and the dose rate at the terminal bud. Results and Discussion: The dose estimation indicated that the radionuclides in the terminal bud and bud scale contributed to the absorbed dose rate mainly by beta rays, whereas those in 1-year-old trunk/branches and leaves were contributed by gamma rays. However, the dose contribution from radionuclides in the lower trunk/branches and leaves was negligible. Conclusion: The fir tree model provides organ-specific DC values, which are satisfactory for the practical calculation of the absorbed dose rate of radiation from inside the tree. These calculations are based on the measurement of radionuclide concentrations in tree organs on the 1-year-old leader shoots of fir trees. With the addition of direct gamma ray measurements of the absorbed dose rate from the tree environment, the total absorbed dose rate was estimated in the terminal bud of fir trees in contaminated forests.

본 논문은 교통망에서 관측 링크 교통량, 미관측 링크의 이용자평형 정보를 이용하여 O-D행렬을 수학적으로 생성하는 모형을 제시하고 있다. 교통량이 관측되지 않은 링크로부터 이용자 평형 상태에서 추출 가능한 정보를 바탕으로 일련의 논리적 연산을 거쳐 실제교통량에 근접하는 서브알고리듬을 유추하여 O-D행렬 추정의 정확도와 연산의 일관성을 제고하였다. 이를 위해 이용자평형상태에서 새로운 정리(Theorem)와 보조정리(Lemma)를 유도하여 적용하였다. 모형의 시험은 3개의 초기 O-D 행렬과 3개의 미관측 링크 교통량 시나리오를 각각의 모형에 적용하여 그 결과를 비교하였다. 적용 결과 본 논문에서 제시된 모형은 기존의 이용자균형 접근방식의 모형emf에 비해 추정된 O-D값의 실제 값과의 차이(O-D Trip RMSE)가 현저히 감소되는 것을 확인하였다.

This research addresses the problem of estimating Origin-Destination (O-D) trip matrices from link volume counts, a set of unobserved link volumes and information of user equilibrium flows in transportation networks. A heuristic algorithm for estimating unobserved link flows is derived, which provides volume estimates that are approximately consistent with both observed flows and an assumption of user equilibrium conditions. These estimated link volumes improve the constraints associated with the synthetic OD estimation model, providing improved solution search procedure. Model performance is tracked in terms of the root mean square errors (RMSE) in predicted travel demands, and where appropriate, predicted linked volumes. These results indicate that the new model substantially outperforms existing approaches to estimating user-equilibrium based synthetic O-D matrices.

Understanding an accurate trip demand by purpose is crucial for short-term regional planning but also long-term regional planning. In traditional approach, information of purpose-oriented trip demand has been derived from public survey in South Korea such as Travel Diary Survey. This type of data acquiring method maybe useful in a sense that it can capture a meaningful sample regarding to entire country, meanwhile it costs a tremendous amount of budget and time. In this research, we want to offer a novel framework for estimating purposeoriented trip demand with dynamic and effective fashions using data fusion in conjunction with Machine learning techniques. With primary results of this concept, this study showed how several state-of-the-art algorithms, including Deep neural network, UMAP, and random forest in conjunction with Genetic algorithm and Tabu-search for optimization, can contribute to this framework. Although tangible results are yet to come, we expect this framework can contribute to resilience planning such as COVID-19.

Formative modeling has been widely used as an alternative to reflective modeling in IS and behavioral research. However, several recent studies have raised questions about its reliability and the research community is divided into its proponents and opponents. Existing studies that attempted the validity testing of a formative model have primarily focused on its estimation stability in the context of varying relationships between a formative construct and the choice of dependent constructs (consequences). In this research, we examine the reliability issue through the lens of a formative construct’s indicators. More specifically, the focus is to study implications of collinearity among formative measures on indicator weight, construct error variance, and structural path coefficient of a formative model. For this, we first frame theoretical expectations that associate the properties of collinearity with the three validity dimensions. Then empirical tests are conducted based on a real survey-based covariance matrix provided by Diamantopoulus (2011) and the results are matched against theoretical expectations to judge the stability of formative modeling.

플라스틱 사출 제품은 다양한 가전제품과 하이테크 제품에 널리 사용되고 있다. 그러나 현재의 치열한 경쟁적 비즈니스 환경에서 플라스틱 사출 제품 제조업자들은 고객을 만족시키면서 경쟁력을 얻기 위하여 다른 경쟁자들보다 먼저 새로운 제품을 시장에 출시하고 신제품의 개발기간을 줄이기 위한 노력을 할 여유가 부족하다. 따라서 무한 경쟁의 시장에서 살아남기 위해서는 제조업자들은 시장 마켓 점유를 빠르게 올리는 것과 동시에 제품의 가격 경쟁력을 가져야 한다. 특징기반 모델의 구조는 현재 연구에서 3D 제작 도구로서 일반적으로 적용되고 있으며 신제품 개발 엔지니어들이 새로운 제품의 개념을 개발하는 데에도 널리 사용되고 있다. 본 연구에서는 특징기반 플라스틱 사출제품을 위한 유전자 알고리즘과 Support Vector Regression (SVR) 기반의 새로운 하이브리드 비용 평가 모델을 제안한다. 제안하는 하이브리드 모델은 기존의 플라스틱 사출제품의 비용평가절차와 계산을 위해 필요로 하는 변수들을 극적으로 간단하게 하고 줄일 수 있다. 사례연구에서는 제안하는 하이브리드 모델과 기존의 multilayer perceptron networks (MLP) 및 pure SVR과의 비교분석을 통하여 제안모델이 플라스틱 사출 제품의 개발단계에서의 비용평가문제를 해결하는데 효율성과 효과성이 있음을 입증한다.

This paper investigates deep learning-based SNR estimation for OFDM systems. A lightweight ResNet-inspired model is applied to estimate SNR under AWGN, Rayleigh, and Rician channels. Specifically, our model consists of two residual blocks to ensure a lightweight design. The dataset includes wide SNR ranges with realistic impairments such as fading and frequency offsets. Performance is evaluated using mean square error (MSE) and mean absolute error (MAE). Results show stable estimation across all channels with low error values in the low SNR regions.

수질 예측은 수질 환경 관리에서 가장 중요한 요소로, 최근 다양한 기계 학습 및 딥러닝 기법을 활용한 연구가 활발히 이루어지고 있다. 환경부에서 유기물질 측정 지표를 CODmn에서 TOC로 전환하는 등 TOC에 대한 중요성이 증대됨에 따라 본 연구는 인천광역시 연수구 승기천을 대상으로 TOC 농도 예측 모델 생성을 위한 최적 입력 변수 및 예측 모델을 제안하는 것을 목표로 한다. 최적 입력 변수 선정은 Pearson 상관도 분석과 Exhaustive Search를 통해 도출하였으며, 최적 예측 모델은 회귀분석, 다층 퍼셉트 론(MLP) 인공신경망, 랜덤포레스트 모델의 정확도를 비교하였다. 모델 성능 비교 결과 T-N, DTN, NH3-N, BOD, NO3-N을 입력 변수로 사용하는 MLP 모델이 가장 우수한 성능을 보였으며, R² 값 0.6280과 RMSE 0.8656의 결과를 기록하였다. 본 연구는 승기천의 TOC 농도 예측을 위한 최적의 입력 변수를 제안함과 동시에, 앞서 언급한 다섯 가지 주요 수질 인자를 활용한 MLP 모델이 유사한 특성을 가진 다른 하천에서 도 TOC 예측에 높은 신뢰도를 제공할 수 있다는 가능성을 제시한다.

Water quality prediction is a key factor in water environment management, with growing interest in utilizing machine learning and deep learning techniques. As the Ministry of Environment transitions the organic matter measurement indicator from CODmn to TOC, the importance of TOC has significantly increased. This study aims to propose optimal input variables and a prediction model for TOC concentration in Seunggi- Stream, Yeonsu-gu, Incheon. Input variables were derived using Pearson correlation analysis and Exhaustive Search. The prediction models tested include Regression Analysis, Multi-Layer Perceptron (MLP) Neural Networks, and Random Forest. Among them, the MLP model using T-N, DTN, NH3-N, BOD, and NO3-N as input variables showed the best performance, achieving an R² of 0.6280 and an RMSE of 0.8656. This study identifies optimal input variables for TOC prediction and suggests that the MLP model can provide reliable TOC predictions for streams with similar characteristics.

본 연구는 물 부족 위기관리를 위한 WEF Nexus(물-에너지-식량 연계성) 관련 기초자료의 확보를 위해, 기후인자와 사회⋅경제요인이 투입된 물 수요모형을 구축하고 국내 용수 이용량에 대한 에너 지 및 식량 공급의 영향력을 실증 분석하였다. 1998년부터 2014년까지 17년간의 정량 분석 자료를 바탕으로, 전국 16개 시도별 용수 이용량과 물 연계요인(에너지소비량, 식량작물생산량), 기후요인 (강수량, 기온), 경제요인(지역내총생산, 수도요금) 관련 자료를 활용하였으며, 지역 간 이질성이 반 영된 가능일반화최소자승(feasible generalized least squares) 패널 모형을 통해 용수 이용량의 결정요 인과 탄력성을 산출하였다. 주요 결과는 다음과 같다. 에너지 소비량, 식량작물생산량, 지역내총생 산, 수도요금 모두 유의한 영향력을 나타냈으나, 강수량과 기온의 효과는 비유의한 것으로 확인되었 다. 물 이용량에 대한 탄력성은 에너지소비량=0.084, 식량작물생산량=0.412, GDRP=0.318, 수도요금 =-0.199로 나타났다.

The aim of this study was to obtain basic information related to WEF (water-energy-food) Nexus for water shortage crisis management by analyzing the influence of energy and food supply on domestic water demand. This study set up a panel model reflecting regional heterogeneity, using the data from 16 metropolitan cities in Korea during the past 17 years (1998 to 2014). The model included multiple variables such as water nexus factors (energy consumption and food crop production), climate factors (precipitation and temperature), and economic factors (GRDP and water rate). The FGLS (feasible generalized least squares) model was selected as the optimal option as a result of the model fitness test. The model results indicate that energy consumption, food crop production, GRDP and water rate all showed significant influence on water demand, but precipitation and temperature were not effective. The elasticity of water consumption was 0.084 for energy consumption, 0.412 for food crops, 0.318 for GDRP, and -0.199 for water rate.

As the requirement on the quality of the system has increased, the reliability is very important part in terms of enhance stability and to provide high quality services to customers. Many statistical models have been developed in the past years for the estimation of software reliability. We consider the functions for NHPP software reliability model and time series model in software failure data. We estimate parameters for the proposed models from three data sets. The values of SSE and MSE is presented from three data sets. We compare the predicted number of faults with the actual three data sets using the NHPP software reliability model and time series model.