• Journal of Agricultural Extension & Community Development
• /
• v.16 no.4
• /
• pp.939-965
• /
• 2009

The Machine Learning has been identified as a promising approach to knowledge-based system development. This study aims to examine the ability of machine learning techniques for farmer's decision making and to develop the reference model for using pig farm data. We compared five machine learning techniques: logistic regression, decision tree, artificial neural network, k-nearest neighbor, and ensemble. All models are well performed to predict the sow's productivity in all parity, showing over 87.6% predictability. The model predictability of total litter size are highest at 91.3% in third parity and decreasing as parity increases. The ensemble is well performed to predict the sow's productivity. The neural network and logistic regression is excellent classifier for all parity. The decision tree and the k-nearest neighbor was not good classifier for all parity. Performance of models varies over models used, showing up to 104% difference in lift values. Artificial Neural network and ensemble models have resulted in highest lift values implying best performance among models.

• Communications for Statistical Applications and Methods
• /
• v.22 no.6
• /
• pp.675-683
• /
• 2015

In the 21st century, we now face the serious problems of the enormous consumption of the energy resources. Depending on the power consumption increases, both China and South Korea face a reduction in available resources. This paper considers the regression models and time-series models to compare the performance of the forecasting accuracy based on Mean Absolute Percentage Error (MAPE) in order to forecast the electricity demand accurately on the short-term period (68 months) data in Northeast China and find the relationship with Korea. Among the models the support vector regression (SVR) model shows superior performance than time-series models for the short-term period data and the time-series models show similar results with the SVR model when we use long-term period data.

• The Korean Journal of Applied Statistics
• /
• v.20 no.2
• /
• pp.219-227
• /
• 2007

The time series models have been used to analyze and predict the network traffic. In this paper, we compare the performance of the time series models for prediction of network traffic. The feasibility study showed that a class of nonlinear time series models can be outperformed than the linear time series models to predict the network traffic.

• Structural Engineering and Mechanics
• /
• v.52 no.1
• /
• pp.97-113
• /
• 2014

In the present work appropriate concrete material models have been proposed to predict drying shrinkage and specific creep of High-performance concrete (HPC) using Artificial Neural Network (ANN). The ANN models are trained, tested and validated using 106 different experimental measured set of data collected from different literatures. The developed models consist of 12 input parameters which include quantities of ingredients namely ordinary Portland cement, fly ash, silica fume, ground granulated blast-furnace slag, water, and other aggregate to cement ratio, volume to surface area ratio, compressive strength at age of loading, relative humidity, age of drying commencement and age of concrete. The Feed-forward backpropagation networks with Levenberg-Marquardt training function are chosen for proposed ANN models and same implemented on MATLAB platform. The results shows that the proposed ANN models are more rational as well as computationally more efficient to predict time-dependent properties of drying shrinkage and specific creep of HPC with high level accuracy.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.6
• /
• pp.1531-1541
• /
• 2013

This paper proposes a new method of VaR forecasting using the conditional autoregressive VaR (CAViaR) models and information criteria. Instead of using a single CAViaR model, we propose to utilize several candidate CAViaR models during a forecasting period. By adopting the Akaike and Bayesian information criteria for quantile regression, we can update not only parameter estimates but also the CAViaR specifications. We also propose extended CAViaR models with a constant location parameter. An empirical study is provided to examine the performance of the proposed method. The results suggest that our method shows more stable performance than those using a single specification.

• Communications for Statistical Applications and Methods
• /
• v.31 no.1
• /
• pp.143-154
• /
• 2024

This paper considers a combined model of exponential smoothing (ETS) and autoregressive integrated moving average (ARIMA) models that are commonly used to forecast time series data. The combined model is constructed through an innovational state space model based on the level variable instead of the differenced variable, and the identifiability of the model is investigated. We consider the maximum likelihood estimation for the model parameters and suggest the model selection steps. The forecasting performance of the model is evaluated by two real time series data. We consider the three competing models; ETS, ARIMA and the trigonometric Box-Cox autoregressive and moving average trend seasonal (TBATS) models, and compare and evaluate their root mean squared errors and mean absolute percentage errors for accuracy. The results show that the combined model outperforms the competing models.

• Korean Journal of Remote Sensing
• /
• v.33 no.4
• /
• pp.445-454
• /
• 2017

The performance of spatial downscaling models depends on the quality of input coarse scale products. Thus, the impact of intrinsic errors contained in coarse scale satellite products on predictive performance should be properly assessed in parallel with the development of advanced downscaling models. Such an assessment is the main objective of this paper. Based on a synthetic satellite precipitation product at a coarse scale generated from rain gauge data, two synthetic precipitation products with different amounts of error were generated and used as inputs for spatial downscaling. Geographically weighted regression, which typically has very high explanatory power, was selected as the trend component estimation model, and area-to-point kriging was applied for residual correction in the spatial downscaling experiment. When errors in the coarse scale product were greater, the trend component estimates were much more susceptible to errors. But residual correction could reduce the impact of the erroneous trend component estimates, which improved the predictive performance. However, residual correction could not improve predictive performance significantly when substantial errors were contained in the input coarse scale data. Therefore, the development of advanced spatial downscaling models should be focused on correction of intrinsic errors in the coarse scale satellite product if a priori error information could be available, rather than on the application of advanced regression models with high explanatory power.

• Korean Journal of Agricultural Science
• /
• v.45 no.4
• /
• pp.859-870
• /
• 2018

Dried red peppers are a staple agricultural product used in Korean cuisine and as such, are an important aspect of agricultural producers' income. Correctly forecasting both their supply and demand situations and price is very important in terms of the producers' income and consumer price stability. The primary objective of this study was to compare the performance of time series forecasting models for dried red peppers in Korea. In this study, three models (an autoregressive model with exogenous variables [ARX], AR-exponential generalized autoregressive conditional heteroscedasticity [EGARCH], and ARX-EGARCH) are presented for forecasting the wholesale price of dried red peppers. As a result of the analysis, it was shown that the ARX model and ARX-EGARCH model, each of which adopt both the rolling window and the adding approach and use the agricultural cooperatives price as the exogenous variable, showed a better forecasting performance compared to the autoregressive model (AR)-EGARCH model. Based on the estimation methods and results, there was no significant difference in the accuracy of the estimation between the rolling window and adding approach. In the case of dried red peppers, there is limitation in building the price forecasting models with a market-structured approach. In this regard, estimating a forecasting model using only price data and identifying the forecast performance can be expected to complement the current pricing forecast model which relies on market shipments.

• Smart Structures and Systems
• /
• v.29 no.1
• /
• pp.207-220
• /
• 2022

Structural health monitoring (SHM) plays a vital role in the maintenance and operation of constructions. In recent years, autonomous inspection has received considerable attention because conventional monitoring methods are inefficient and expensive to some extent. To develop autonomous inspection, a potential approach of crack identification is needed to locate defects. Therefore, this study exploits two deep learning-based segmentation models, DeepLabv3+ and Mask R-CNN, for crack segmentation because these two segmentation models can outperform other similar models on public datasets. Additionally, impacts of label quality on model performance are explored to obtain an empirical guideline on the preparation of image datasets. The influence of image cropping and label refining are also investigated, and different strategies are applied to the dataset, resulting in six alternated datasets. By conducting experiments with these datasets, the highest mean Intersection-over-Union (mIoU), 75%, is achieved by Mask R-CNN. The rise in the percentage of annotations by image cropping improves model performance while the label refining has opposite effects on the two models. As the label refining results in fewer error annotations of cracks, this modification enhances the performance of DeepLabv3+. Instead, the performance of Mask R-CNN decreases because fragmented annotations may mistake an instance as multiple instances. To sum up, both DeepLabv3+ and Mask R-CNN are capable of crack identification, and an empirical guideline on the data preparation is presented to strengthen identification successfulness via image cropping and label refining.

• KIEAE Journal
• /
• v.14 no.2
• /
• pp.3-10
• /
• 2014

Impact by estimation error of hourly horizontal global solar radiation in a weather file on building energy performance was investigated in this study. There are a number of weather parameters in a given weather file, such as dry-bulb, wet-bulb, dew-point temperatures; wind speed and direction; station pressure; and solar radiation. Most of them except for solar radiation can be easily obtained from weather stations located on the sites worldwide. However, most weather stations, also including the ones in South Korea, do not measure solar radiation because the measuring equipment for solar radiation is expensive and difficult to maintain. For this reason, many researchers have studied solar radiation estimation models and suggested to apply them to predict solar radiation for different weather stations in South Korea, where the solar radiation is not measured. However, only a few studies have been conducted to identify the impact caused by estimation errors of various solar radiation models on building energy performance analysis. Therefore, four different weather files using different horizontal global solar radiation data, one using measured global solar radiation, and the other three using estimated global solar radiation models, which are Cloud-cover Radiation Model (CRM), Zhang and Huang Model (ZHM), and Meteorological Radiation Model (MRM) were packed into TRY formatted weather files in this study. These were then used for office building energy simulations to compare their energy consumptions, and the results showed that there were differences in the energy consumptions due to these four different solar radiation data. Additionally, it was found that using hourly solar radiation from the estimation models, which had a similar hourly tendency with the hourly measured solar radiation, was the most important key for precise building energy simulation analysis rather than using the solar models that had the best of the monthly or yearly statistical indices.