• Atmosphere
• /
• v.31 no.1
• /
• pp.73-83
• /
• 2021

Numerical weather prediction (NWP) models play an essential role in predicting weather factors, but using them is challenging due to various factors. To overcome the difficulties of NWP models, deep learning models have been deployed in weather forecasting by several recent studies. This study adapts long short-term memory (LSTM), which demonstrates remarkable performance in time-series prediction. The combination of LSTM model input of meteorological features and activation functions have a significant impact on the performance therefore, the results from 5 combinations of input features and 4 activation functions are analyzed in 9 Automated Surface Observing System (ASOS) stations corresponding to cities/islands/mountains. The optimized LSTM model produces better performance within eight forecast hours than Local Data Assimilation and Prediction System (LDAPS) operated by Korean meteorological administration. Therefore, this study illustrates that this LSTM model can be usefully applied to very short-term weather forecasting, and further studies about CNN-LSTM model with 2-D spatial convolution neural network (CNN) coupled in LSTM are required for improvement.

• The Journal of Asian Finance, Economics and Business
• /
• v.10 no.2
• /
• pp.49-60
• /
• 2023

The study attempts to examine the relationship among revenue growth factors from different angles and provides a comprehensive overview of tax revenue collection for developing countries. The impact of income tax, customs duty, and value-added tax on the gross domestic product is examined using the ordinary least-square (OLS) multiple regression approach. To confirm the association, a multiple regression model is applied to time-series data. SPSS software, MS Excel, is used to draw the empirical results, trend analysis, and some graphical presentation to reach the study's objective. The findings show that while the value-added tax has a significant impact and the highest coefficient, regardless of country, income tax and customs duty may or may not be significant depending on the circumstances. It triggers effectual and efficacious economic growth. The paper has implications in policy-making areas where governments are seeking how to stimulate revenue growth effectively and efficiently. To promote economic growth, the tax net and tax rate on luxury goods should be increased along with human resources in the tax administration for the short term. But in the long term, decentralization & digitization of tax administration, dismantling the existing tax barriers and good governance are necessary.

• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.7-13
• /
• 2021

This study develops an artificial intelligence prediction system for Fine particulate Matter(PM2.5) based on the deep learning algorithm GAN model. The experimental data are closely related to the changes in temperature, humidity, wind speed, and atmospheric pressure generated by the time series axis and the concentration of air pollutants such as SO2, CO, O3, NO2, and PM10. Due to the characteristics of the data, since the concentration at the current time is affected by the concentration at the previous time, a predictive model for recursive supervised learning was applied. For comparative analysis of the accuracy of the existing models, CNN and LSTM, the difference between observation value and prediction value was analyzed and visualized. As a result of performance analysis, it was confirmed that the proposed GAN improved to 15.8%, 10.9%, and 5.5% in the evaluation items RMSE, MAPE, and IOA compared to LSTM, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.453-459
• /
• 2021

The system marginal price of electricity is the amount paid to all the generating units, which is an important decision-making factor for the construction and maintenance of an electrical power unit. In this paper, we suggest a long-term forecasting model for calculating the system marginal price based on prices of natural gas and oil. As most variables used in the analysis are nonstationary time series, the long run relationship among the variables should be examined by cointegration tests. The forecasting model is similar to an error correction model which consists of a long run cointegrating equation and another equation for short run dynamics. To mitigate the robustness issue arising from the relatively small data sample, this study employs various testing and estimating methods. Compared to previous studies, this paper considers multiple fuel prices in the forecasting model of system marginal price, and provides greater emphasis on the robustness of analysis. As none of the cointegrating relations associated with system marginal price, natural gas price and oil price are excluded, three error correction models are estimated. Considering the root mean squared error and mean absolute error, the model based on the cointegrating relation between system marginal price and natural gas price performs best in the out-of-sample forecast.

• The Journal of Asian Finance, Economics and Business
• /
• v.6 no.3
• /
• pp.163-172
• /
• 2019

The paper aims to investigate the impact of the gap between domestic and international gold price on money demand in Vietnam, an emerging economy in the Asian region. We use a quarterly database collected from the first quarter of 2004 to the fourth quarter of 2016. The time-series database includes 52 observations. The money demand is represented by M2; Domestic income is the Gross domestic product at the constant prices of 1994; Inflation rate is calculated by the Customer Price Index from the General Statistics Office of Vietnam. The result confirms the existence of a long-term cointegration relationship between the money demand and the gap between domestic and international gold price as well as some variables including domestic income, inflation, and real exchange rate. The regression results also show that the gap between domestic and international gold price has a positive impact on money demand in the Vietnamese economy. Besides, the domestic income and international gold price have positive impacts on money demand while the inflation and real exchange rate are negatively related in the long run. This proves that the gap between the domestic and international gold price really has a positive impact on money demand in Vietnam during the study period.

• Imaging Science in Dentistry
• /
• v.49 no.4
• /
• pp.307-315
• /
• 2019

This report describes 3 cases of cleidocranial dysplasia (CCD) and presents relevant findings on long-term follow-up radiographic images of impacted permanent teeth with delayed eruption. Radiographic images of 3 CCD patients were reviewed retrospectively. These images were mainly composed of panoramic and skull radiographs, and the follow-up periods were 3, 13, and 13 years, respectively. The distinct features revealed by the images were described, and the eruption state of impacted permanent teeth was evaluated. The features common to the 3 cases were multiple supernumerary teeth, the presence of Wormian bone, underdevelopment of the maxilla and the maxillary sinus, and clavicular hypoplasia. The eruption of impacted permanent teeth was not observed without proper dental treatment in adult CCD cases, even after long time periods had elapsed. When proper orthodontic force was applied, tooth movement was observed in a manner not significantly different from the general population.

• The Bulletin of The Korean Astronomical Society
• /
• v.38 no.2
• /
• pp.70.2-70.2
• /
• 2013

We present collections of optical photometry for a pre-main sequence star HBC 722. It showed large amplitude optical outburst (${\Delta}V=4.7$ mag) in 2010 and classified as a FU Orionis type object. We have been observing HBC 722 from 2011 April to 2013 May, using Camera for QUasars in EArly uNiverse (CQUEAN) attached to the 2.1 m Otto Struve telescope at the McDonald Observatory. Time-series monitoring data (minute-scale interval) were obtained in SDSS r, i and z bands to see short-scale behaviors as well as trace the long-term brightness changes after the eruption in 2010. Interestingly, it started to brighten from 2011 early summer and became brighter than the first outburst peak in our 2013 May observation. We expect that the recovering phase would result from re-increase of disk accretion rate, might attribute to distinctive short-scale color features. In this presentation, we report long- and short-timescale optical behaviors of HBC 722 in the post-outburst phase.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.1400-1402
• /
• 2003

The objective of this paper is to determine the suitable spectral bands for monitoring growth status change during a long period. The long-term ground-level reflectance spectra as well as LAI and biomass were obtained in xiaotangshan area, Beijing, 2001. The narrow-band NDVI type spectral indices by all possible two bands were calculated their correlation coefficients R$^2$ with biomass and LAI. The best NDVIs must have higher R$^2$ with both biomass and LAI. The reasonable band centers and band widths were determined by a systematically increasing bandwidth centered over a wavelength. In addition, the first 19 bands of MODIS were simulated and investigated. Each developed spectral indices was then validated by the biomass and LAI time series using the generalized vector angle. It turned out that six new NDVI type indices within 750-1400nm were developed. NDVI(811_10,957_10) and NDVI(962_10,802_10) performed best. No satisfactory conventional NDVI formed by red and NIR bands were found effective. MODIS_NDVI(band19, band17) and MODIS_NDVI(band19, band2) were much better than MODIS_NDVI(band2,band1) for growth monitoring.

• Geomechanics and Engineering
• /
• v.5 no.1
• /
• pp.37-55
• /
• 2013

An important issue in the design of soil-nailing systems, as long-term retaining walls, is to assess their stability during seismic events. As such, this study is aimed at simulating the dynamic behavior and failure pattern of nailed structures using two series of numerical analyses, namely dynamic time history and pseudo-static. These numerical simulations are performed using the Finite Difference Method (FDM). In order to consider the actual response of a soil-nailed structure, nonlinear soil behaviour, soil-structure interaction effects, bending resistance of structural elements and construction sequences have been considered in the analyses. The obtained results revealed the efficiency of both analysis methods in simulating the seismic failure mechanism. The predicted failure pattern consists of two sliding blocks enclosed by three slip surfaces, whereby the bottom nails act as anchors and the other nails hold a semi-rigid soil mass. Moreover, it was realized that an increase in the length of the lowest nails is the most effective method to improve seismic stability of soil-nailed structures. Therefore, it is recommended to first estimate the nails pattern for static condition with the minimum required static safety factor. Then, the required seismic stability can be obtained through an increase in the length of the lowest nails. Moreover, placement of additional long nails among lowest nails in existing nailed structures can be considered as a simple retrofitting technique in seismic prone areas.

• Asia pacific journal of information systems
• /
• v.30 no.4
• /
• pp.719-740
• /
• 2020

The stock market changes continuously as new information emerges, affecting the judgments of investors. Online news articles are valued as a traditional window to inform investors about various information that affects the stock market. This paper proposed new ways to utilize online news articles with technical indicators. The suggested hybrid model consists of three models. First, a self-attention-based convolutional neural network (CNN) model, considered to be better in interpreting the semantics of long texts, uses news content as inputs. Second, a self-attention-based, bi-long short-term memory (bi-LSTM) neural network model for short texts utilizes news titles as inputs. Third, a bi-LSTM model, considered to be better in analyzing context information and time-series models, uses 19 technical indicators as inputs. We used news articles from the previous day and technical indicators from the past seven days to predict the share price of the next day. An experiment was performed with Korean stock market data and news articles from 33 top companies over three years. Through this experiment, our proposed model showed better performance than previous approaches, which have mainly focused on news titles. This paper demonstrated that news titles and content should be treated in different ways for superior stock price prediction.