• 대한인간공학회지
• /
• 제28권3호
• /
• pp.55-62
• /
• 2009

This study is concerned with predicting human search performance using a visual lobe. The most previous studies on human performance in visual search have been limited to a single-target search. This study extended the visual search research to multiple-target search including targets of different types as well as targets of same types. A model for predicting visual search performance was proposed and the model was validated by human search data. Additionally, this study found that human subjects always did not use a constant ratio of the whole visual lobe size for each type of targets in visual search process. The more conspicuous the target is, the more ratio of the whole visual lobe size human subjects use. The model that can predict human performance in multiple-target search may facilitate visual inspection plan in manufacturing.

• 공업화학
• /
• 제31권5호
• /
• pp.520-525
• /
• 2020

본 연구에서는 증류공정의 제품 생산단 온도 예측을 위한 머신러닝 기반 모델을 개발하였다. 증류공정의 제어는 제품 생산단의 온도를 통해 이루어지고 있어 제어를 위해 정확한 온도 예측이 필요하다. 증류공정에서 온도는 다양한 변수들과 복잡한 비선형의 관계를 형성하고 있으며 시계열 데이터의 특성이 있어 이를 예측하기 위해 순환신경망 기반 알고리즘을 이용하였다. 모델 개발 과정에서 적절한 예측 알고리즘을 선정하기 위해 세 가지 순환신경망 기반 알고리즘과 배치 사이즈 조절하여 제품 생산단 온도를 예측하기 위한 가장 적저한 모델을 선정하였다. LSTM₁₂₈ 모델이 제품 생산단 온도를 예측하기 위한 가장 적절한 모델로 선정되었다. 선정된 모델을 활용하여 실제 공정 운전데이터에 적용한 결과 RMSE 0.0791, R² 0.924의 성능을 보였다.

• 부동산연구
• /
• 제27권3호
• /
• pp.71-86
• /
• 2017

본 연구의 목적은 딥 러닝 방법을 부동산가격지수 예측에 적용해보고, 기존의 시계열분석 방법과의 비교를 통해 부동산 시장 예측의 새로운 방법으로서 활용가능성을 확인하는 것이다. 딥 러닝(deep learning)방법인 DNN(Deep Neural Networks)모형 및 LSTM(Long Shot Term Memory networks)모형과 시계열분석 방법인 ARIMA(autoregressive integrated moving average)모형을 이용하여 여러 가지 부동산가격지수에 대한 예측을 시도하였다. 연구결과 첫째, 딥 러닝 방법의 예측력이 시계열분석 방법보다 우수한 것으로 나타났다. 둘째, 딥 러닝 방법 중에서는 DNN모형의 예측력이 LSTM모형의 예측력보다 우수하나 그 정도는 미미한 수준인 것으로 나타났다. 셋째, 딥 러닝 방법과 ARIMA모형은 부동산 가격지수(real estate price index) 중 아파트 실거래가격지수(housing sales price index)에 대한 예측력이 가장 부족한 것으로 나타났다. 향후 딥 러닝 방법을 활용함으로써 부동산 시장에 대한 예측의 정확성을 제고할 수 있을 것으로 기대된다.

• 대기
• /
• 제17권2호
• /
• pp.135-145
• /
• 2007

We developed a simple regression model for predicting the intesity change (central pressure) of major tropical cyclones (TCs) for 24 hours after landfall using 51 TC cases which landed over the Korean Peninsula for 1951-2004. Clusters 1 and 4 with a relatively strong intensity of TC after landfall classified by Choi and Kim (2007) are used to develop a statistical model for the prediction of TC intensity change. Predicting parameters (falling constants) in the regression models $(P_t=P_0+alnt)$ are 6.46 and 10.11 for clusters 1 and 4, respectively. It might be mentioned that there is some feasibility in employing a simple regression model developed in this study for TC intensity change after landfall for operational purpose of TC forecasting compared with RSMC-Tokyo best-track in both TC cases of Clusters 1 and 4 and Ewiniar (0603) case, but the room for improvement of model still remains for further study.

• 한국환경과학회지
• /
• 제26권4호
• /
• pp.421-432
• /
• 2017

In this study, we examined the suitability of ten disinfection models for predicting the inactivation of Artemia sp. via single or combined physical and chemical treatments. The effect of Hydraulic Retention Time (HRT) on the inactivation of Artemia sp. was examined experimentally. Disinfection models were fitted to the experimental data by using the GInaFiT plug-in for Microsoft Excel. The inactivation model were evaluated on the basis of RMSE (Root Mean Square Error), SSE (mean Sum Square Error) and $r^2$. An inactivation model with the lowest RMSE, SSE and $r^2$ close to 1 was considered the best. The Weibull+Tail model was found to be the most appropriate for predicting the inactivation of Artemia sp. via electrolytic treatment and electrolytic-ultrasonic combined treatment. The Log-linear+Tail model was the most appropriate for modeling inactivation via homogenization and combined electrolytic-homogenization treatment. The double Weibull disinfection model was the most suitable for the predicting inactivation via ultrasonic treatment.

• 한국경영과학회지
• /
• 제34권3호
• /
• pp.29-41
• /
• 2009

The purpose of this paper is to develop a new classification method for predicting consumer choice based on genetic algorithm, and to validate Its prediction power over existing methods. To serve this purpose, we propose a hybrid model, and discuss Its methodological characteristics in comparison with other existing classification methods. Also, we conduct a series of experiments employing survey data of consumer choices of MP3 players to assess the prediction power of the model. The results show that the suggested model in this paper is statistically superior to the existing methods such as logistic regression model, artificial neural network model and decision tree model in terms of prediction accuracy. The model is also shown to have an advantage of providing several strategic information of practical use for consumer choice.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2008년도 추계학술대회 및 정기총회
• /
• pp.1-7
• /
• 2008

The purpose of this paper is to develop a new classification method for predicting consumer choice based on genetic algorithm, and to validate its prediction power over existing methods. To serve this purpose, we propose a hybrid model, and discuss its methodological characteristics in comparison with other existing classification methods. Also, to assess the prediction power of the model, we conduct a series of experiments employing survey data of consumer choices of MP3 players. The results show that the suggested model in this paper is statistically superior to the existing methods such as logistic regression model, artificial neural network model and decision tree model in terms of prediction accuracy. The model is also shown to have an advantage of providing several strategic information of practical use for consumer choice.

• 대한가정학회지
• /
• 제43권11호
• /
• pp.225-234
• /
• 2005

The objective of this study was to develop a model for predicting Korean adult consumers who frequently eat food-away-from-home. A total of 7,032 adults aged 19 years and older from the 2001 National Health and Nutrition Survey in Korea were used as subjects. The data were analyzed using a data mining procedure including logistic regression and decile analysis. The model developed in the study was proven to be valid in predicting the consumers who frequently eat food-away-from home(once a day or more often). This model showed that consumers eating food-away-from-home frequently tend to be younger men, living in a big city, working full time, receiving more stress and eating snacks and fried food more frequently. The model could be used to identify targets for nutrition and related education and consumer segments for the marketing of restaurant businesses.

• Journal of Electrical Engineering and Technology
• /
• 제13권3호
• /
• pp.1089-1098
• /
• 2018

This study aims at developing and applying a hybrid model to the wind power prediction (WPP). The hybrid model for a very-short-term WPP (VSTWPP) is achieved through analytical data, multiple linear regressions and least square methods (MLR&LS). The data used in our hybrid model are based on the historical records of wind power from an offshore region. In this model, the WPP is achieved in four steps: 1) transforming historical data into ratios; 2) predicting the wind power using the ratios; 3) predicting rectification ratios by the total wind power; 4) predicting the wind power using the proposed rectification method. The proposed method includes one-step and multi-step predictions. The WPP is tested by applying different models, such as the autoregressive moving average (ARMA), support vector machine (SVM), and artificial neural network (ANN). The results of all these models confirmed the validity of the proposed hybrid model in terms of error as well as its effectiveness. Furthermore, forecasting errors are compared to depict a highly variable WPP, and the correlations between the actual and predicted wind powers are shown. Simulations are carried out to definitely prove the feasibility and excellent performance of the proposed method for the VSTWPP versus that of the SVM, ANN and ARMA models.

• Computers and Concrete
• /
• 제33권3호
• /
• pp.285-299
• /
• 2024

Recently, many engineering computations have realized their digital transformation to Machine Learning (ML)-based systems. Predicting the behavior of a structure, which is mainly computed with structural analysis software, is an essential step before construction for efficient structural analysis. Especially in the seismic-based design procedure of the structures, predicting the lateral load capacity of reinforced concrete (RC) columns is a vital factor. In this study, a novel ML-based model is proposed to predict the maximum lateral load capacity of RC columns under varying axial loads or cyclic loadings. The proposed model is generated with a Deep Neural Network (DNN) and compared with traditional ML techniques as well as a popular commercial structural analysis software. In the design and test phases of the proposed model, 319 columns with rectangular and square cross-sections are incorporated. In this study, 33 parameters are used to predict the maximum lateral load capacity of each RC column. While some traditional ML techniques perform better prediction than the compared commercial software, the proposed DNN model provides the best prediction results within the analysis. The experimental results reveal the fact that the performance of the proposed DNN model can definitely be used for other engineering purposes as well.