A causality exists between insect density and plant health, where plant health is affected by both the plant’s potential and environmental factors. In other words, causality is possible between insect density and environmental factors, allowing for the analysis of insect density based on these environmental factors. Machine learning enables studying insect density alongside environmental factors, providing insights into the causality between insects, the environment, and plant health. Machine learning is a methodology that involves the design of models by learning patterns from input data. This study aims to predict F. occidentalis density by sampling environmental factors and applying them to machine learning models.

Recently, as the possibility of unexpected outbreaks of alien insects has increased due to climate change such as global warming, the importance of early control through rapid and accurate spread of exotic forest pest and change prediction diagnosis is required. This study summarizes and reports the followings: the establishment of monitoring strategy for exotic insects by the investigation of species distribution range through field surveys and others, the development of new diagnostic technique through microstructures and life-cycle, the dispersal of exotic insects, and ecological impact assessment using ecological methods and with the expansion of exotic insects and development of ecosystem impact prediction model.

Chloride is one of the most common threats to reinforced concrete (RC) durability. Alkaline environment of concrete makes a passive layer on the surface of reinforcement bars that prevents the bar from corrosion. However, when the chloride concentration amount at the reinforcement bar reaches a certain level, deterioration of the passive protection layer occurs, causing corrosion and ultimately reducing the structure’s safety and durability. Therefore, understanding the chloride diffusion and its prediction are important to evaluate the safety and durability of RC structure. In this study, the chloride diffusion coefficient is predicted by machine learning techniques. Various machine learning techniques such as multiple linear regression, decision tree, random forest, support vector machine, artificial neural networks, extreme gradient boosting annd k-nearest neighbor were used and accuracy of there models were compared. In order to evaluate the accuracy, root mean square error (RMSE), mean square error (MSE), mean absolute error (MAE) and coefficient of determination (R2) were used as prediction performance indices. The k-fold cross-validation procedure was used to estimate the performance of machine learning models when making predictions on data not used during training. Grid search was applied to hyperparameter optimization. It has been shown from numerical simulation that ensemble learning methods such as random forest and extreme gradient boosting successfully predicted the chloride diffusion coefficient and artificial neural networks also provided accurate result.

목적 : 본 연구의 목적은 Coronavirys Disease(COVID-19) 팬데믹을 거친 고령자의 사회적 건강의 수준의 변화궤적에 대한 잠재계층을 분류하고 잠재 계층별 특성을 분석함에 있다. 또한 이러한 사회적 건강에 영향을 미치는 예측요인을 파악하여 고령자의 사회적 건강을 증진을 위한 기초자료를 마련하고자 한다. 연구방법 : 국내 고령자의 사회적 친밀도에 따른 사회적 건강 유형을 파악하기 위해서 한국복지패널의 3 년차 종단자료를 토대로 분석하였으며, 연구대상자는 세 시점 모두를 응답한 2845명의 고령자를 대상으 로 하였다. 대상자중심접근인 성장혼합모형(Growth Mixture Model; GMM)을 적용하여 변화궤적에 따 른 잠재계층을 분석 하였고, 도출된 각 잠재유형별 특성을 파악하기 위해 χ2 분석, 분산분석을 실시하 였으며, 계층 간 차이를 유발하는 요인을 파악하기 위해 다항로지스틱 회귀분석을 실시하였다. 결과 : GMM 적용결과, 사회적 건강의 변화궤적에 대한 잠재계층은 최종 4개의 집단으로 저수준 감소-증 가 집단, 중수준 유지-증가 집단, 고수준-감소 집단, 고수준 유지’집단으로 분류되었다. 또한 사회적 건강 수준에 따라 여가만족도에서 차이가 나타나는 것으로 드러났으며, 그 외에도 연령 차이가 존재하였 다. 잠재계층분류에 영향을 미치는 영향변인을 검증한 결과, 특히 여성일수록, 종교를 가지고 있을수록, 여가만족도와 전반적 만족도가 모두 높을수록 고수준 유지 집단에 속할 확률이 높은 것으로 나타났다. 결론 : 국내 고령자의 사회적 건강은 시간이 지남에 따라 감소하는 궤적을 보이는 것으로 나타났다. 변화 궤적에 따라 4개의 집단으로 구분 지을 수 있으며, 각 잠재 유형별 연령과 여가 만족도 부분에서 집단별 차이가 드러났다.

본 논문에서는 소형어선의 운동 응답을 예측하기 위해 딥러닝 모델을 구축하였다. 크기가 다른 두 소형어선을 대상으로 유체 동역학 성능을 평가하여 데이터세트를 확보하였다. 딥러닝 모델은 순환 신경망 기법의 하나인 장단기 메모리 기법(LSTM, Long Short-Term Memory)을 사용하였다. 딥러닝 모델의 입력 데이터는 6 자유도 운동 및 파고의 시계열 데이터를 사용하였으며, 출력 라벨로는 6 자유도 운동의 시계열 데이터로 선정하였다. 최적 LSTM 모델 구축을 위해 hyperparameter 및 입력창 길이의 영향을 평가하였다. 구축된 LSTM 모 델을 통해 입사파 방향에 따른 시계열 운동 응답을 예측하였다. 예측된 시계열 운동 응답은 해석 결과와 전반적으로 잘 일치함을 확인 할 수 있었다. 시계열의 길이가 길어짐에 따라서 예측값과 해석 결과의 차이가 발생하는데, 이는 장기 데이터에 따른 훈련 영향도가 감 소 됨에 따라 나타난 것으로 확인할 수 있다. 전체 예측 데이터의 오차는 약 85% 이상의 데이터가 10% 이내의 오차를 보였으며, 소형어 선의 시계열 운동 응답을 잘 예측함을 확인하였다. 구축된 LSTM 모델은 소형어선의 모니터링 및 경보 시스템에 활용될 수 있을 것으로 기대한다.

본 논문에서는 볼트로 체결된 구조체에 대하여 초기 볼트풀림 상태에서의 볼트 체결력 예측 합성곱 신경망 훈련 방법을 제시한다. 8개의 볼트의 체결력이 변경된 상태에서 계산한 주파수응답들을 완전 체결된 상태의 초기 모델과의 크기 및 모양 유사성을 표현하는 유사성 지도로 생성한다. 주파수응답 데이터들의 생성에는 크리로프 부공간법 기반의 모델차수축소법을 적용하여 효율적인 방법으 로 수행할 수 있도록 한다. 합성곱 신경망 모델은 회귀 출력 계층을 사용하여 볼트의 체결력을 예측하도록 하였으며, 훈련 데이터의 개 수와 합성곱 신경망 계층의 개수를 다르게 준비하여 훈련시킨 네트워크들을 비교하여 그 성능을 평가하였다. 주파수응답에서 파생되 는 유사성 지도를 입력 데이터로 사용하여 초기 볼트풀림 영역에서 볼트 체결력의 진단 가능성과 유효성을 제시하였다.

본 연구에서는 대파의 가락시장 도매가격을 이용하여 기존 시계열 모형인 ARIMA 모형, 홀트-윈터스 평활법과 대표적인 기계학습 방법인 랜덤 포레스트(Random forest) 분석 기법의 가격 예측력을 비교하였다. 세 모형의 예측력을 분석한 결과는 다음과 같다. 가장 예측력이 높게 나타난 모형은 3년(36개월)을 주기로 설정한 ARIMA 모형이었다. 또한 ARIMA 모형과 홀트-윈터스 평활법은 일별 데이터보다 월별 데이터를 이용한 예측 결과의 정확도가 더 높아 훈련 데이터에 대한 과적합(overfitting)이 오히려 예측력을 낮추는 현상을 보였다. 반면, 랜덤 포레스트는 월별 데이터 보다 일별 데이터를 사용한 모형의 예측력이 더 높았다. 이는 학습량이 많을수록 높은 예측력을 보여주는 기계학습의 특징을 보여주었다. 그러나 기계학습 방법을 활용한 가격 예측에는 가격에 영향을 주는 설명변수를 찾고, 양질의 훈련 데이터 축적이 필요하다는 것을 알 수 있었다. 향후 연구에서는 다양한 설명변수와 기계학습 및 딥러닝 기법을 적용한다면 농축산물 가격 예측력을 높이는데 도움이 될 것으로 판단된다.

PURPOSES : Due to the frequent occurrence of accidents on icy roads during nighttime, it would be advantageous to notify road managers and drivers about the most perilous areas. This would allow road managers to treat the icy roads with de-icing chemicals and enable drivers to be better prepared for potential hazards. Essential information about pavement temperature is required to identify icy spots on the road. METHODS : With the goal of estimating nighttime pavement temperature on the National Highways in Korea using atmospheric data, the current study investigated a widely recognized forecasting method known as deep neural network (DNN). To achieve this objective, the input data for the models were gathered from the weather agency's website. The dataset comprised of relative humidity, air temperature, dew point temperature, as well as the differences in air temperature and humidity between two consecutive days. RESULTS : In order to assess the effectiveness of the built DNN model, a comparison was made using baseline pavement temperature data gathered through an infrared-based pavement temperature sensor installed in a highway patrol car. The results indicated that the DNN model achieved a mean absolute error (MAE) of 0.42 and a root mean square error (RMSE) of 0.62. In comparison, a conventional regression model yielded an MAE of 2.07 and an RMSE of 2.64. Thus, the DNN model demonstrated superior performance in comparison to the conventional regression model. CONCLUSIONS : Considering the increasing focus on preventive maintenance, these newly developed prediction models can be implemented proactively as a preventive measure against icing. This proactive approach has the potential to significantly improve traffic safety on winter roads.

In this study, based on the saturation magnetic flux density experimental values (Bs) of 622 Fe-based bulk metallic glasses (BMGs), regression models were applied to predict Bs using artificial neural networks (ANN), and prediction performance was evaluated. Model performance evaluation was investigated by using the F1 score together with the coefficient of determination (R2 score), which is mainly used in regression models. The coefficient of determination can be used as a performance indicator, since it shows the predicted results of the saturation magnetic flux density of full material datasets in a balanced way. However, the BMG alloy contains iron and requires a high saturation magnetic flux density to have excellent applicability as a soft magnetic material, and in this study F1 score was used as a performance indicator to better predict Bs above the threshold value of Bs (1.4 T). After obtaining two ANN models optimized for the R2 and F1 score conditions, respectively, their prediction performance was compared for the test data. As a case study to evaluate the prediction performance, new Fe-based BMG datasets that were not included in the training and test datasets were predicted using the two ANN models. The results showed that the model with an excellent F1 score achieved a more accurate prediction for a material with a high saturation magnetic flux density.

수소는 다양한 신재생에너지 중 환경친화적인 에너지로 각광받고 있지만 농업에 적용된 사례는 드물다. 본 연구는 수소 연료전지 삼중 열병합 시스템을 온실에 적용하여 에너지를 절 약하고 온실가스를 줄이고자 한다. 이 시스템은 배출된 열을 회수하면서 수소로부터 난방, 냉각 및 전기를 생산할 수 있다. 수소 연료 전지 삼중 열 병합 시스템을 온실에 적용하기 위해 서는 온실의 냉난방 부하 분석이 필요하다. 이를 위해서는 온 실의 형태, 냉난방 시스템, 작물 등을 고려해야 한다. 따라서 본 연구에서는 건물 에너지 시뮬레이션(BES)을 활용하여 냉 난방 부하를 추정하고자 한다. 전주지역의 토마토를 재배하 는 반밀폐형 온실을 대상으로 2012년부터 2021년까지의 기 상데이터를 수집하여 분석했다. 온실 설계도를 참고하여 피 복재와 골조를 모델화하여 작물 에너지와 토양 에너지 교환을 실시했다. 건물 에너지 시뮬레이션의 유효성을 검증하기 위 해 작물의 유무에 의한 분석, 정적 에너지 및 동적 에너지 분석 을 실시했다. 또한 월별 최대 냉난방 부하 분석에 의해 평균 최 대 난방 용량 449,578kJ·h-1, 냉방 용량 431,187kJ·h-1이 산정 되었다.

Consumers' online reviews have become more powerful in the Internet market. Consumers share reviews, post comments and constantly evaluate products online. In previous studies, the analysis of online reviews mainly focused on purchasing products based on consumers' own use experience, but in innovative products, it was difficult to find an analysis of product acceptor's response to product user reviews. In particular, there is no online review study of VR covered in this study. This study not only quantitatively analyzed online reviews of consumers who purchased VR products on Amazon, an online distribution site, but also qualitatively analyzed them through crawling. This study used Amazon's VR product user review, where purchases were confirmed, to select algorithms that are more likely to be matched by predicting a helpful review and presenting a predictive model. In addition, the online review extracted deep text associated with Helpful and conducted topical modeling. As a result, topics related to 1) experience in use, 2) post-product evaluation, 3) product composition and peripherals, 4) immersion, and 5) comfort were highly acceptable to potential inmates. To enhance the acceptability of innovative products through online reviews, it is not just highlighting the product advantages of VR, but also suggests that the link between smartphones and applications can bring in more potential users. Also, interworking with other peripheral devices (speakers or screens) can be predicted as a way to increase the acceptability of VR products. From a marketing perspective, this study has found targeted topics that help consumers in pioneering the VR market, which will help potential customers create the services they want.

We propose a sales prediction model based on the number of new members, online advertising, and consumer reviews for a short period. Considering purchase behaviors of new and existing members, we predict reliable sales amounts, which can be monthly updated. Our study provides digital marketers with a feasible prediction approach.

In the design of HLW repositories, it is important to confirm the performance and safety of buffer materials at high temperatures. Most existing models for predicting hydraulic conductivity of bentonite buffer materials have been derived using the results of tests conducted below 100°C. However, they cannot be applied to temperatures above 100°C. This study suggests a prediction model for the hydraulic conductivity of bentonite buffer materials, valid at temperatures between 100°C and 125°C, based on different test results and values reported in literature. Among several factors, dry density and temperature were the most relevant to hydraulic conductivity and were used as important independent variables for the prediction model. The effect of temperature, which positively correlates with hydraulic conductivity, was greater than that of dry density, which negatively correlates with hydraulic conductivity. Finally, to enhance the prediction accuracy, a new parameter reflecting the effect of dry density and temperature was proposed and included in the final prediction model. Compared to the existing model, the predicted result of the final suggested model was closer to the measured values.

The prolonged and heavy East Asian summer precipitation in 2020 may have been caused by an enhanced Madden-Julian Oscillation (MJO), which requires evaluation using forecast models. We examined the performance of GloSea6, an operational forecast model, in predicting the East Asian summer precipitation during July 2020, and investigated the role of MJO in the extreme rainfall event. Two experiments, CON and EXP, were conducted using different convection schemes, 6A and 5A, respectively to simulate various aspects of MJO. The EXP runs yielded stronger forecasts of East Asian precipitation for July 2020 than the CON runs, probably due to the prominent MJO realization in the former experiment. The stronger MJO created stronger moist southerly winds associated with the western North Pacific subtropical high, which led to increased precipitation. The strengthening of the MJO was found to improve the prediction accuracy of East Asian summer precipitation. However, it is important to note that this study does not discuss the impact of changes in the convection scheme on the modulation of MJO. Further research is needed to understand other factors that could strengthen the MJO and improve the forecast.