• The Journal of Asian Finance, Economics and Business
• /
• 제8권3호
• /
• pp.751-761
• /
• 2021

The Vietnamese economy is a developing country that has brought many opportunities and challenges for the banking system. Commercial banks have developed strongly from quality to quantity, which plays a vital role in developing the economy. They play an important role in capital formation, which is essential for the economic development of a country. They provide financial services to the general public and businesses, ensuring economic and social stability and sustainable growth of the economy. Therefore, the relationship between bank depth and economic growth is of importance in research. This paper used a VAR (Vector Autoregressive Models) estimator for time series data models. The data is collected quarterly from the first quarter of the year 2000 to 2020. The study uses the VAR model to examine the causal relationships of economic growth, growth in money supply expansion, private sector capital requirement, and banks' domestic credit. The results indicate a general short-run relationship between banking sector depth and economic growth with a positive connection, but in the long term, the relationship between these variables can be reversed because of other macro factors. The findings show the two-way causal relationship between GDP growth and banking depth factors. This research contributes to policy-making by underlining the banking sector depth determinants when setting regulations and policies to develop the banking sector.

• Asia-Pacific Journal of Atmospheric Sciences
• /
• 제54권4호
• /
• pp.599-610
• /
• 2018

The goal of this study was to statistically analyse the variability of global irradiance and ultraviolet erythemal (UVER) irradiance and their interrelationships with global and UVER irradiance, global clearness indices and ozone. A prediction of short-term UVER solar irradiance values was also obtained. Extreme values of UVER irradiance were included in the data set, as well as a time series of ultraviolet irradiance variability (UIV). The study period was from 2005 to 2014 and approximately 250,000 readings were taken at 5-min intervals. The effect of the clearness indices on global irradiance variability (GIV) and UIV was also recorded and bi-dimensional distributions were used to gather information on the two measured variables. With regard to daily GIV and UIV, it is also shown that for global clearness index ($k_t$) values lower than 0.6 both global and UVER irradiance had greater variability and that UIVon cloud-free days ($k_t$ higher than 0.65) exceeds GIV. To study the dependence between UIVand GIV the ${\chi}^2$ statistical method was used. It can be concluded that there is a 95% probability of a clear dependency between the variabilities. A connection between high $k_t$ (corresponding to cloudless days) and low variabilities was found in the analysis of bidimensional distributions. Extreme values of UVER irradiance were also analyzed and it was possible to calculate the probable future values of UVER irradiance by extrapolating the values of the adjustment curve obtained from the Gumbel distribution.

• 한국항만경제학회지
• /
• 제37권3호
• /
• pp.1-17
• /
• 2021

본 연구에서는 항만의 단기 물동량을 예측하기 위해 ARIMA 모형과 CART 모형을 활용한 단기 수요예측 모형을 제시하였다. 제시한 모형은 2단계로 구성된다. 1단계에서는 시계열 예측치와 주요 교역국의 주당 근로일수를 변수로 사용하여 CART 모형을 추정하고 주별 물동량 예측을 진행한다. 2단계에서는 1단계에서 도출한 예측치와 요일 정보, 주요국 공휴일 정보, 주요국 행사 기간 정보를 설명변수로 활용하여 최종적인 일별 물동량 예측 모형을 추정한다. 제시한 수요예측 모형을 활용하여 2020년 10월 1일부터 12월 31일까지 92일의 부산항 물동량을 예측한 결과 제시한 모형의 평균 정확도가 기존 시계열 모형보다 '22.5%' 높은 것으로 나타났다. 제시 모형은 일별 물동량의 추세뿐만 아니라 물동량이 급등락하는 지점에서도 높은 정확도를 보였으며 시계열 예측 모형을 사용했을 때 비해 총 166,504(TEU)의 오차를 줄일 수 있는 것으로 나타났다. 항만의 효율적인 운영을 위해 필수적인 단기 물동량 예측에 적합한 예측 모형을 제시한 본 연구는 충분한 활용 가치가 있을 것으로 판단된다.

• 인터넷정보학회논문지
• /
• 제24권1호
• /
• pp.39-47
• /
• 2023

The development of autonomous driving and Advanced Driver Assistance System (ADAS) technology has grown rapidly in recent years. As most traffic accidents occur due to human error, self-driving vehicles can drastically reduce the number of accidents and crashes that occur on the roads today. Obviously, technical advancements in autonomous driving can lead to improved public driving safety. However, due to the current limitations in technology and lack of public trust in self-driving cars (and drones), the actual use of Autonomous Vehicles (AVs) is still significantly low. According to prior studies, people's acceptance of an AV is mainly determined by trust. It is proven that people still feel much more comfortable in personalized ADAS, designed with the way people drive. Based on such needs, a new attempt for a customized ADAS considering each driver's driving style is proposed in this paper. Each driver's behavior is divided into two categories: assertive and defensive. In this paper, a novel customized ADAS algorithm with high classification accuracy is designed, which divides each driver based on their driving style. Each driver's driving data is collected and simulated using CARLA, which is an open-source autonomous driving simulator. In addition, Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) machine learning algorithms are used to optimize the ADAS parameters. The proposed scheme results in a high classification accuracy of time series driving data. Furthermore, among the vast amount of CARLA-based feature data extracted from the drivers, distinguishable driving features are collected selectively using Support Vector Machine (SVM) technology by comparing the amount of influence on the classification of the two categories. Therefore, by extracting distinguishable features and eliminating outliers using SVM, the classification accuracy is significantly improved. Based on this classification, the ADAS sensors can be made more sensitive for the case of assertive drivers, enabling more advanced driving safety support. The proposed technology of this paper is especially important because currently, the state-of-the-art level of autonomous driving is at level 3 (based on the SAE International driving automation standards), which requires advanced functions that can assist drivers using ADAS technology.

• 농업과학연구
• /
• 제49권2호
• /
• pp.193-202
• /
• 2022

The recurrent neural network (RNN) algorithm has been widely used in water-related research areas, such as water level predictions and water quality predictions, due to its excellent time series learning capabilities. However, studies on water quality predictions using RNN algorithms are limited because of the scarcity of water quality data. Therefore, most previous studies related to water quality predictions were based on monthly predictions. In this study, the quality of the water in a reservoir in Nonsan, Chungcheongnam-do Republic of Korea was predicted using the RNN-LSTM algorithm. The study was conducted after constructing data that could then be, linearly interpolated as daily data. In this study, we attempt to predict the water quality on the 7th, 15th, 30th, 45th and 60th days instead of making daily predictions of water quality factors. For daily predictions, linear interpolated daily water quality data and daily weather data (rainfall, average temperature, and average wind speed) were used. The results of predicting water quality concentrations (chemical oxygen demand [COD], dissolved oxygen [DO], suspended solid [SS], total nitrogen [T-N], total phosphorus [TP]) through the LSTM algorithm indicated that the predictive value was high on the 7th and 15th days. In the 30th day predictions, the COD and DO items showed R² that exceeded 0.6 at all points, whereas the SS, T-N, and T-P items showed differences depending on the factor being assessed. In the 45th day predictions, it was found that the accuracy of all water quality predictions except for the DO item was sharply lowered.

• Journal of Animal Science and Technology
• /
• 제65권4호
• /
• pp.792-803
• /
• 2023

This study aimed to assess the impact of weather events on the sorghum-sudangrass hybrid (Sorghum bicolor L.) cultivar production trend in the central inland region of Korea during the monsoon season, using time series analysis. The sorghum-sudangrass production data collected between 1988 and 2013 were compiled along with the production year's weather data. The growing degree days (GDD), accumulated rainfall, and sunshine duration were used to assess their impacts on forage production (kg/ha) trend. Conversely, GDD and accumulated rainfall had positive and negative effects on the trend of forage production, respectively. Meanwhile, weather events such as heavy rainfall and typhoon were also collected based on weather warnings as weather events in the Korean monsoon season. The impact of weather events did not affect forage production, even with the increasing frequency and intensity of heavy rainfall. Therefore, the trend of forage production for the sorghum-sudangrass hybrid was forecasted to slightly increase until 2045. The predicted forage production in 2045 will be 14,926 ± 6,657 kg/ha. It is likely that the damage by heavy rainfall and typhoons can be reduced through more frequent harvest against short-term single damage and a deeper extension of the root system against soil erosion and lodging. Therefore, in an environment that is rapidly changing due to climate change and extreme/abnormal weather, the cultivation of the sorghum-sudangrass hybrid would be advantageous in securing stable and robust forage production. Through this study, we propose the cultivation of sorghum-sudangrass hybrid as one of the alternative summer forage options to achieve stable forage production during the dynamically changing monsoon, in spite of rather lower nutrient value than that of maize (Zea mays L.).

• 한국컴퓨터정보학회논문지
• /
• 제29권7호
• /
• pp.73-80
• /
• 2024

본 연구는 토픽 모델링과 장단기 기억(LSTM) 신경망을 결합하여 한국 종합주가지수(KOSPI) 예측의 정확도를 향상하는 방법을 제안한다. 본 논문에서는 LDA(Latent Dirichlet Allocation) 기법을 이용해 금융 뉴스 데이터에서 금리 인상 및 인하와 관련된 10개의 주요 주제를 추출하고, 추출된 주제를 과거 KOSPI 지수와 함께 LSTM 모델에 입력하여 KOSPI 지수를 예측하는 모델을 제안한다. 제안된 모델은 과거 KOSPI 지수를 LSTM 모델에 입력하여 시계열 예측 방법과 뉴스 데이터를 입력하여 토픽 모델링하는 방법을 결합하여 KOSPI 지수를 예측하는 특성을 가진다. 제안된 모델의 성능을 검증하기 위해, 본 논문에서는 LSTM의 입력 데이터의 종류에 따라 4개의 모델(LSTM_K 모델, LSTM_KNS 모델, LDA_K 모델, LDA_KNS 모델)을 설계하고 각 모델의 예측 성능을 제시하였다. 예측 성능을 비교한 결과, 금융 뉴스 주제 데이터와 과거 KOSPI 지수 데이터를 입력으로 하는 LSTM 모델(LDA_K 모델)이 가장 낮은 RMSE(Root Mean Square Error)를 기록하여 가장 좋은 예측 성능을 보였다.

• 한국측량학회지
• /
• 제37권3호
• /
• pp.199-208
• /
• 2019

운용 가능한 위성의 수가 증가하고 기술이 진보함에 따라 영상정보의 성과물이 다양해지고 많은 양의 자료가 축적되고 있다. 본 연구에서는 기구축된 영상정보를 활용하여 부족한 훈련자료의 문제를 극복하고 딥러닝(deep learning) 기법의 장점을 활용하고자 전이학습과 변화탐지 네트워크를 활용한 고해상도 위성영상의 변화탐지를 수행하였다. 본 연구에서 활용한 딥러닝 네트워크는 공간 및 분광 정보를 추출하는 합성곱 레이어(convolutional layer)와 시계열 정보를 분석하는 합성곱 장단기 메모리 레이어(convolutional long short term memory layer)로 구성되었으며, 고해상도 다중분광 영상에 최적화된 정보를 추출하기 위하여 커널(kernel)의 차원에 따른 정확도를 비교하였다. 또한, 학습된 커널 정보를 활용하기 위하여 변화탐지 네트워크의 초기 합성곱 레이어를 고해상도 항공영상인 ISPRS (International Society for Photogrammetry and Remote Sensing) 데이터셋에서 추출된 40,000개의 패치로 학습된 값으로 초기화하였다. 다시기 KOMPSAT-3A (KOrean Multi-Purpose SATllite-3A) 영상에 대한 실험 결과, 전이학습과 딥러닝 네트워크를 활용할 경우 기복 변위 및 그림자 등으로 인한 변화에 덜 민감하게 반응하며 분류 항목이 달라진 지역의 변화를 보다 효과적으로 추출할 수 있었으며, 2차원 커널보다 3차원 커널을 사용할 때 변화탐지의 정확도가 높았다. 3차원 커널은 공간 및 분광정보를 모두 고려하여 특징 맵(feature map)을 추출하기 때문에 고해상도 영상의 분류뿐만 아니라 변화탐지에도 효과적인 것을 확인하였다. 본 연구에서는 고해상도 위성영상의 변화탐지를 위한 전이학습과 딥러닝 기법의 활용 가능성을 제시하였으며, 추후 훈련된 변화탐지 네트워크를 새롭게 취득된 영상에 적용하는 연구를 수행하여 제안기법의 활용범위를 확장할 예정이다.

• Ocean and Polar Research
• /
• 제37권2호
• /
• pp.127-140
• /
• 2015

A continuous series of 60 snow samples was collected at a 2.5-cm interval from a 1.5-m snow pit at a site on the Styx Glacier Plateau in Victoria Land, Antarctica, during the 2011/2012 austral summer season. Various chemical components (${\delta}D$, ${\delta}^{18}O$, $Na^+$, $K^+$, $Mg^{2+}$, $Ca^{2+}$, $Cl^-$, $SO_4{^2-}$, $NO_3{^-}$, $F^-$, $CH_3SO_3{^-}$, $CH_3CO_2{^-}$ and $HCO_2{^-}$) were determined to understand the highly resolved seasonal variations of these species in the coastal atmosphere near the Antarctic Jang Bogo station. Based on vertical profiles of ${\delta}^{18}O$, $NO_3{^-}$and MSA, which showed prominent seasonal changes in concentrations, the snow samples were dated to cover the time period from 2009 austral winter to 2012 austral summer with a mean accumulation rate of $226kgH_2Om^{-2}yr^{-1}$. Our snow profiles show pronounced seasonal variations for all the measured chemical species with a different pattern between different species. The distinctive feature of the occurrence patterns of the seasonal variations is clearly linked to changes in the relative strength of contributions from various natural sources (sea salt spray, volcanoes, crust-derived dust, and marine biogenic activities) during different short-term periods. The results allow us to understand the transport pathways and input mechanisms for each species and provide valuable information that will be useful for investigating long-term (decades to century scale periods) climate and environmental changes that can be deduced from an ice core to be retrieved from the Styx Glacier Plateau in the near future.

• 수산해양교육연구
• /
• 제28권2호
• /
• pp.323-335
• /
• 2016

The main reason of marine casualties is the human error in respect of ship's operation. The human error of officers in charge of a navigational watch is related to their abilities to handle of navigational equipment. Navigational devices play a key role to help officers decide what to do for safe navigation. Thus, the abilities to handle of navigational equipment mean not only operation of devices but also entire understanding of the system such as interpretation of information obtained from devices, appropriate use of information considering navigational circumstance. Qualification of seafarers is in accordance with STCW and detailed training courses for their qualification are provided by IMO model course series. Recently, ships engaged on international voyages shall be fitted with an ECDIS not later than the first survey on or after 1 July 2018. As increasing use of ECDIS on ships, marine casualties related to ECDIS are on the rise. The primary causes of the accidents are lacking understanding of ECDIS system, wrong presentation of information on display, wrong safety setting by seafarers who use ECDIS, using small-scale chart and missing charts update. As a result of these primary causes, some problems like wrong route planning and use of limited or omitted information occur. It could be happening by inappropriate seafarers' training which is not sufficient to support improving abilities of officers to handle navigational equipment. For efficient training, it is need to develop training courses. Applying full mission simulation system to seafarers' training courses with case studies and best practices which are well-constructed scenarios based on true marine casualties can increase the effect of training. To use the simulation system, it is possible that seafarers are trained under condition that closely resemble real situation. It should be considered that IMO model course be revised depending on the level of seafarers also. It could be helpful for increasing seafarers' abilities of equipment operation in place of accumulation of experience spending much time. In the short term, effort of training courses improvement for seafarers is needed and long term, it should be tried to provide stable system and services relate to ECDIS.