• 농업과학연구
• /
• 제48권4호
• /
• pp.681-691
• /
• 2021

The estimated breeding value (EBV) and accuracy of Hanwoo steer (Korean cattle) is an indicator that can predict the slaughter time in the future and carcass performance outcomes. Recently, studies using pedigrees and genotypes are being actively conducted to improve the accuracy of the EBV. In this study, the pedigree and genotype of 46 steers obtained from livestock farm A in Gyeongnam were used for a pedigree best linear unbiased prediction (PBLUP) and a genomic best linear unbiased prediction (GBLUP) to estimate and analyze the breeding value and accuracy of the carcass weight (CWT), eye muscle area (EMA), back-fat thickness (BFT), and marbling score (MS). PBLUP estimated the EBV and accuracy by constructing a numeric relationship matrix (NRM) from the 46 steers and reference population I (545,483 heads) with the pedigree and phenotype. GBLUP estimated genomic EBV (GEBV) and accuracy by constructing a genomic relationship matrix (GRM) from the 46 steers and reference population II (16,972 heads) with the genotype and phenotype. As a result, in the order of CWT, EMA, BFT, and MS, the accuracy levels of PBLUP were 0.531, 0.519, 0.524 and 0.530, while the accuracy outcomes of GBLUP were 0.799, 0.779, 0.768, and 0.810. The accuracy estimated by GBLUP was 50.1 - 53.1% higher than that estimated by PBLUP. GEBV estimated with the genotype is expected to show higher accuracy than the EBV calculated using only the pedigree and is thus expected to be used as basic data for genomic selection in the future.

• 한국환경보건학회지
• /
• 제50권2호
• /
• pp.113-124
• /
• 2024

Background: The increasing need to minimize animal testing has sparked interest in alternative methods with more humane, cost-effective, and time-saving attributes. In particular, in silico-based computational toxicology is gaining prominence. Adverse outcome pathway (AOP) is a biological map depicting toxicological mechanisms, composed of molecular initiating events (MIEs), key events (KEs), and adverse outcomes (AOs). To understand toxicological mechanisms, predictive models are essential for AOP components in computational toxicology, including molecular structures. Objectives: This study reviewed the literature and investigated previous research cases related to AOP and in silico methodologies. We describe the results obtained from the analysis, including predictive techniques and approaches that can be used for future in silico-based alternative methods to animal testing using AOP. Methods: We analyzed in silico methods and databases used in the literature to identify trends in research on in silico prediction models. Results: We reviewed 26 studies related to AOP and in silico methodologies. The ToxCast/Tox21 database was commonly used for toxicity studies, and MIE was the most frequently used predictive factor among the AOP components. Machine learning was most widely used among prediction techniques, and various in silico methods, such as deep learning, molecular docking, and molecular dynamics, were also utilized. Conclusions: We analyzed the current research trends regarding in silico-based alternative methods for animal testing using AOPs. Developing predictive techniques that reflect toxicological mechanisms will be essential to replace animal testing with in silico methods. In the future, since the applicability of various predictive techniques is increasing, it will be necessary to continue monitoring the trend of predictive techniques and in silico-based approaches.

• Structural Engineering and Mechanics
• /
• 제91권1호
• /
• pp.39-47
• /
• 2024

Bridge deterioration shows the change of bridge condition during its operation, and predicting bridge deterioration is important for implementing predictive protection and planning future maintenance. However, in practical application, the raw inspection data of bridges are not continuous, which has a greater impact on the accuracy of the prediction results. Therefore, two kinds of bridge deterioration models are established in this paper: one is based on the traditional regression theory, combined with the distribution fitting theory to preprocess the data, which solves the problem of irregular distribution and incomplete quantity of raw data. Secondly, based on the theory of Long Short-Term Memory (LSTM) Recurrent Neural Network (RNN), the network is trained using the raw inspection data, which can realize the prediction of the future deterioration of bridges through the historical data. And the inspection data of 60 prestressed concrete box girder bridges in Xiamen, China are used as an example for validation and comparative analysis, and the results show that both deterioration models can predict the deterioration of prestressed concrete box girder bridges. The regression model shows that the bridge deteriorates gradually, while the LSTM-RNN model shows that the bridge keeps great condition during the first 5 years and degrades rapidly from 5 years to 15 years. Based on the current inspection database, the LSTM-RNN model performs better than the regression model because it has smaller prediction error. With the continuous improvement of the database, the results of this study can be extended to other bridge types or other degradation factors can be introduced to improve the accuracy and usefulness of the deterioration model.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제12권3호
• /
• pp.125-132
• /
• 2023

본 연구에서는 식물의 영양분 흡수에 따른 식물 성장뿐만 아니라 기공 기능 및 광합성에도 영향을 끼치는 온실의 수증기압차(VPD, Vapor Pressure Deficit)예측을 위한 머신러닝 모델들의 성능을 비교해보았다. VPD 예측을 위해 온실 내·외부 환경요소 및 시계열 데이터의 시간적 요소들과의 상관관계를 확인하고 상관관계가 높은 요소들이 VPD에 어떤 영향을 미치는지 확인하였다. 예측 모델의 성능을 분석하기 전 분석 시계열 데이터의 양(1일, 3일, 7일), 간격(20분, 1시간)이 예측 성능에 미치는 영향을 확인하여 데이터의 양과 간격을 조절하였다. 마지막으로 4개의 머신러닝 예측 모델(XGB Regressor, LGBM Regressor, Random Forest Regressor 등)을 적용하여 모델별 예측 성능을 비교했다. 모델의 예측 결과로 20분 간격의 1일의 데이터를 사용했을 때 LGBM에서 MAE는 0.008, RMSE는 0.011의 가장 높은 예측 성능을 보였다. 또한 20분 후 VPD 예측에 가장 큰 영향을 미치는 요소는 환경적 요인보다는 과거 20분 전의 VPD(VPD_y__71)임을 확인하였다. 본 연구의 결과를 활용하여 VPD 예측을 통해 작물의 생산성을 높이고, 온실의 결로, 병 발생 예방 등이 가능하다. 향후 온실의 환경 데이터 예측뿐만 아니라 더 나아가 생산량 예측, 스마트팜 제어 모델 등 다양한 분야에 활용할 수 있을 것이다.

• 한국환경보건학회지
• /
• 제27권3호
• /
• pp.49-56
• /
• 2001

The water quality simulation was carried out to predict water quality in Tan stream of the Han river using water quality model, QUAL2E. In the end, the future variations in water quality of Tan stream were simulated and the prediction of the impacts of Tan stream on water quality in the Han river was carried out by applying the Tan stream simulation results into the model. The results are as follows. The predicted results of future water quality of Tan stream suggested that the concentrations of BOD, T-N and T-P at Chungdam bridge would increase to 0.68~0.77 mg/$\ell$, 1.33~1.62 mg/$\ell$ and 0.05~0.06 mg/$\ell$, respectively in 2006 and 2011 and that with the implementation of advanced treatment in Sungnam and Tanchun sewage treatment plants, the concentration of T-N would be reduced more as the amount of treated sewage increase, while the concentration of T-P would stay 0.49 mg/$\ell$. The results obtained from simulation of the impacts of future Tan stream water quality improvement on the main stream of the Han river showed that with implementation of advanced treatment in both Sungnam and Tanchun sewage treatment plants, the concentration of T-N, T-P and chlorophyll-a at Hangang bridge and Heangju bridge would be reduced by 11.6%, 7.7% and 20.9%, respectively in 2..6 and by 13.6%, 9.4% and 22.2%, respectively in 2011, which indicates that the effect on the reduction of T-N and T-P would be relatively significant while the effect on the decrease of algae would be slight.

• 한국환경복원기술학회지
• /
• 제18권1호
• /
• pp.103-113
• /
• 2015

Korean pines(Pinus koraiensis) are one of the major plantation species in the Republic of Korea and their natural habitats range from Japan and China to Siberia. The seed of Korean pines, pine nuts, are well know for good food reserves. Due to the global changes which drive the Korean peninsula into the semi-tropical climate, current plantations and natural habitats of Korean pines are faced with the change in the environmental conditions to some extent. To anticipate suitable sites for Korean pines in the future, the location of Korean pines were extracted from the 'Map of suitable trees on a site' that provides the map of suitable trees on a site considering tree species for timber and special uses, and then MaxEnt modelling was used for generating a prediction map on the basis of statistical analysis. As a result, the order of predicted suitable sites were Kangwon-do, Kyungsangbuk-do and Chungcheongbuk-do provinces and sites with high elevation within those provinces were considered most suitable in common. The prediction map of suitable sites for Korean pines presented that suitable sites in the future decreased by 72.2% by 2050's and almost disappeared with a decrease of 92.1% by 2070's on a nationwide scale. In relation to the major production regions of pine nuts in South Korea - Gapyung gun and Yangpyung gun, Kyunggi province and Hongcheon gun, Kangwon province, suitable sites within their areas were predicted to increase by 2050's but become extinct in South Korea by 2070's. To establish a long-term countermeasures against the improvement on forest productivity quality in terms of managing national food security, the result from this study can be considered as a firm basis of predicting plantation suitability. Also, it can be used to predict the changes in supply of forest products and thereby market values in accordance with climate change scenarios.

• 전기전자학회논문지
• /
• 제22권3호
• /
• pp.680-692
• /
• 2018

Children's health is interesting to parents and society. A system that assists to monitor the development of their children and gives nutritional advices is an interesting of parents. In this study, we present a system that allows to track the heights and weights of a child since he/she was born up to adulthood, to predict his age of puberty, and to provide nutritional advice. Particularly, it predicts the height in near future and the adult stature for detecting the child with abnormal development. We applied Sager's model for predicting the height in near future by using interpolation and regression techniques before puberty. After determining the puberty time, we proposed a model for predicting the height. Then we applied fuzzy logic for evaluating the health status and providing nutritional advice. Our system predicted stature in near future with error bound of $1.7361{\pm}0.0397cm$ in girls and $2.4020{\pm}0.0799cm$ in boys. Our model also gave a reliable adult stature prediction with error bound of $0.3507{\pm}0.2808cm$ in girls and $1.3414{\pm}0.7024cm$ in boys. At the same time, the nutrition was provided appropriately in terms of protein, lipid, glucid. We implemented a program based on this research. Our system promises to improve the health of every child.

• 한국시뮬레이션학회논문지
• /
• 제25권3호
• /
• pp.65-76
• /
• 2016

자가-적응 소프트웨어는 스스로 문제를 인지하여 인지한 문제에 대하여 소프트웨어 사이클이 멈추지 않고 해당 요구사항에 맞게 적응하는 소프트웨어이다. 본 논문에서는 임계점이 존재하는 시스템에서 발생하는 불필요한 적응 수행을 감소시키기 위하여 선행적 방식으로 임계점 이후의 상황을 예측함으로써 문제가 되는 이벤트를 사전에 처리하고자 한다. 실측치는 대부분 선형과 비선형이 모두 나타나기 때문에 하이브리드 모델을 사용하여 임계점 이후를 예측하며, 예측 기법의 사용 여부는 예측의 정확도를 기반으로 하는 적응 시점 판단 지표를 기준으로 한다. 본 논문의 기여점으로는 하이브리드 모델을 MAPE-K에 적용하여 임계점 이후 상황을 예측함으로써 실제 변화에 대한 불확실성을 감소시켰다는 점과 적응 시점 판단 지표를 기반으로 적응 시점을 판단함으로써 불필요한 적응 수행을 줄였다는 데에 있다.

• 한국전자거래학회지
• /
• 제24권1호
• /
• pp.91-103
• /
• 2019

기업 입장에서 앞으로 있을 현금흐름에 대한 예측이란, 향후 발생할 수 있는 유동성(현금부족) 위험을 미리 파악할 수 있다는 점과 미래의 투자계획을 세우는데 중요한 자료가 될 수 있다는 점에서 중요한 의의를 지닌다. 그러나 기업 간 거래에서 매출 채권 형태로 발생하는 거래 유형은 다른 유형의 거래와는 달리 채무 이행 불확실성이 존재하며, 이로 인해 정확한 현금흐름 예측을 어렵게 한다. 본 연구에서는 추계적 분석 기법의 하나인 가변 마코프 기법(Variable Order Markov model)을 활용하여 기업 간에 발생 할 수 있는 매출 채권과 관련한 현금흐름 동향을 예측한다. 구체적으로는, PST(Probabilistic Suffix Tree)라는 가변 마코프 기법을 활용하여, 지난 과거의 매출 채권 발행 및 수금 내역을 바탕으로 해당 매출 채권들의 기대 연령 예측 연구를 수행하였다. 본 연구결과를 통해, 기존의 다른 기법들과 대비하여 가변 마코프 기법을 활용 시, 평균 12.5% 이상의 정확도를 보여주고 있음을 밝혔다.

• KNOM Review
• /
• 제23권1호
• /
• pp.34-42
• /
• 2020

NFV (Network Function Virtualization) 환경에서는 소프트웨어로 구현된 가상 네트워크 기능 (VNF: Virtualized Network Function)을 범용 서버에 설치하는 것으로 네트워크 기능을 제공한다. 네트워크 관리자는 VNF를 네트워크 토폴로지 상 적절한 위치의 서버에 배치하고 상황에 따라 동적으로 관리함으로써, 다양한 네트워크 상황에 대해 신속하고 유연하게 대응할 수 있다. 하지만 여러 네트워크 조건 (서비스 비용 및 품질) 등을 고려하는 것은 매우 복잡하고 어려운 문제이며, 특히 결정된 배치를 실제 NFV 환경에 적용하는 데는 처리 시간이 소요되기 때문에, 최적의 VNF 배치를 위해서는 필요한 자원량을 예측하여 VNF 배치를 결정하는 것이 필요하다. 본 논문에서는 MEC (Multi-access Edge Computing) 토폴로지에서 서비스 요청을 무작위로 생성하여 ILP (Integer Linear Programming) 모델을 통해 시뮬레이션한 결과를 학습데이터로 사용하는 기계학습 모델을 도출한다. 도출된 예측 모델은 5분 이후의 미래 시점에 대해 ILP 솔루션 결과 대비 90% 이상의 정확도를 보였다.