• Journal of Cadastre & Land InformatiX
• /
• v.47 no.2
• /
• pp.159-174
• /
• 2017

The long-term imbalances between supply and demand in office market can weaken urban growth since excessive supply of offices led to office market instability and excessive demand of offices weakens growth of urban industry. Recently, there have been a lot of new large-scale supplies, which increased volatility in Seoul office market. Nevertheless, new supply of Seoul office has not been fully examined. Given this, the focus of this article was on confirming the influences of profitability, replacement cost, and demand on new office supplies in Seoul. In examining those influences, another focus was on their relative influences over time. For these purposes, we analyzed quarterly data of Seoul office market between 2003 and 2015 using a vector error correction model (VECM). As a result, in terms of the influences on the current new supply, the impact of supply before the first quarter was negative, while that of office employment before the first quarter was positive. Also, that of interest rate before the second quarter was positive, while those of cap rate before the first quarter and cap rate before the second quarter were negative. Based on the findings, it is suggested that prediction models on Seoul offices need to be developed considering the influences of profitability, replacement cost, and demand on new office supplies in Seoul.

• Journal of Intelligence and Information Systems
• /
• v.26 no.4
• /
• pp.127-148
• /
• 2020

The data center is a physical environment facility for accommodating computer systems and related components, and is an essential foundation technology for next-generation core industries such as big data, smart factories, wearables, and smart homes. In particular, with the growth of cloud computing, the proportional expansion of the data center infrastructure is inevitable. Monitoring the health of these data center facilities is a way to maintain and manage the system and prevent failure. If a failure occurs in some elements of the facility, it may affect not only the relevant equipment but also other connected equipment, and may cause enormous damage. In particular, IT facilities are irregular due to interdependence and it is difficult to know the cause. In the previous study predicting failure in data center, failure was predicted by looking at a single server as a single state without assuming that the devices were mixed. Therefore, in this study, data center failures were classified into failures occurring inside the server (Outage A) and failures occurring outside the server (Outage B), and focused on analyzing complex failures occurring within the server. Server external failures include power, cooling, user errors, etc. Since such failures can be prevented in the early stages of data center facility construction, various solutions are being developed. On the other hand, the cause of the failure occurring in the server is difficult to determine, and adequate prevention has not yet been achieved. In particular, this is the reason why server failures do not occur singularly, cause other server failures, or receive something that causes failures from other servers. In other words, while the existing studies assumed that it was a single server that did not affect the servers and analyzed the failure, in this study, the failure occurred on the assumption that it had an effect between servers. In order to define the complex failure situation in the data center, failure history data for each equipment existing in the data center was used. There are four major failures considered in this study: Network Node Down, Server Down, Windows Activation Services Down, and Database Management System Service Down. The failures that occur for each device are sorted in chronological order, and when a failure occurs in a specific equipment, if a failure occurs in a specific equipment within 5 minutes from the time of occurrence, it is defined that the failure occurs simultaneously. After configuring the sequence for the devices that have failed at the same time, 5 devices that frequently occur simultaneously within the configured sequence were selected, and the case where the selected devices failed at the same time was confirmed through visualization. Since the server resource information collected for failure analysis is in units of time series and has flow, we used Long Short-term Memory (LSTM), a deep learning algorithm that can predict the next state through the previous state. In addition, unlike a single server, the Hierarchical Attention Network deep learning model structure was used in consideration of the fact that the level of multiple failures for each server is different. This algorithm is a method of increasing the prediction accuracy by giving weight to the server as the impact on the failure increases. The study began with defining the type of failure and selecting the analysis target. In the first experiment, the same collected data was assumed as a single server state and a multiple server state, and compared and analyzed. The second experiment improved the prediction accuracy in the case of a complex server by optimizing each server threshold. In the first experiment, which assumed each of a single server and multiple servers, in the case of a single server, it was predicted that three of the five servers did not have a failure even though the actual failure occurred. However, assuming multiple servers, all five servers were predicted to have failed. As a result of the experiment, the hypothesis that there is an effect between servers is proven. As a result of this study, it was confirmed that the prediction performance was superior when the multiple servers were assumed than when the single server was assumed. In particular, applying the Hierarchical Attention Network algorithm, assuming that the effects of each server will be different, played a role in improving the analysis effect. In addition, by applying a different threshold for each server, the prediction accuracy could be improved. This study showed that failures that are difficult to determine the cause can be predicted through historical data, and a model that can predict failures occurring in servers in data centers is presented. It is expected that the occurrence of disability can be prevented in advance using the results of this study.

• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.241-254
• /
• 2011

Financial time-series forecasting is one of the most important issues because it is essential for the risk management of financial institutions. Therefore, researchers have tried to forecast financial time-series using various data mining techniques such as regression, artificial neural networks, decision trees, k-nearest neighbor etc. Recently, support vector machines (SVMs) are popularly applied to this research area because they have advantages that they don't require huge training data and have low possibility of overfitting. However, a user must determine several design factors by heuristics in order to use SVM. For example, the selection of appropriate kernel function and its parameters and proper feature subset selection are major design factors of SVM. Other than these factors, the proper selection of instance subset may also improve the forecasting performance of SVM by eliminating irrelevant and distorting training instances. Nonetheless, there have been few studies that have applied instance selection to SVM, especially in the domain of stock market prediction. Instance selection tries to choose proper instance subsets from original training data. It may be considered as a method of knowledge refinement and it maintains the instance-base. This study proposes the novel instance selection algorithm for SVMs. The proposed technique in this study uses genetic algorithm (GA) to optimize instance selection process with parameter optimization simultaneously. We call the model as ISVM (SVM with Instance selection) in this study. Experiments on stock market data are implemented using ISVM. In this study, the GA searches for optimal or near-optimal values of kernel parameters and relevant instances for SVMs. This study needs two sets of parameters in chromosomes in GA setting : The codes for kernel parameters and for instance selection. For the controlling parameters of the GA search, the population size is set at 50 organisms and the value of the crossover rate is set at 0.7 while the mutation rate is 0.1. As the stopping condition, 50 generations are permitted. The application data used in this study consists of technical indicators and the direction of change in the daily Korea stock price index (KOSPI). The total number of samples is 2218 trading days. We separate the whole data into three subsets as training, test, hold-out data set. The number of data in each subset is 1056, 581, 581 respectively. This study compares ISVM to several comparative models including logistic regression (logit), backpropagation neural networks (ANN), nearest neighbor (1-NN), conventional SVM (SVM) and SVM with the optimized parameters (PSVM). In especial, PSVM uses optimized kernel parameters by the genetic algorithm. The experimental results show that ISVM outperforms 1-NN by 15.32%, ANN by 6.89%, Logit and SVM by 5.34%, and PSVM by 4.82% for the holdout data. For ISVM, only 556 data from 1056 original training data are used to produce the result. In addition, the two-sample test for proportions is used to examine whether ISVM significantly outperforms other comparative models. The results indicate that ISVM outperforms ANN and 1-NN at the 1% statistical significance level. In addition, ISVM performs better than Logit, SVM and PSVM at the 5% statistical significance level.

• Journal of Korea Water Resources Association
• /
• v.42 no.7
• /
• pp.579-589
• /
• 2009

Hwang River in South Korea, has experienced channel adjustments due to dam construction. Hapcheon main dam and re-regulation dam. The reach below the re-regulation dam (45 km long) changed in flow regime, channel width, bed material distribution, vegetation expansion, and island formation after dam construction. The re-regulation dam dramatically reduced annual peak flow from 654.7 $m^3$/s to 126.3 $m^3$/s and trapped the annual 591 thousand $m^3$ of sediment load formerly delivered from the upper watershed since the completion of the dam in 1989. An analysis of a time series of aerial photographs taken in 1982, 1993, and 2004 showed that non-vegetated active channel width narrowed an average of 152 m (47% of 1982) and non-vegetated active channel area decreased an average of 6.6 km2 (44% of 1982) between 1982 and 2004, with most narrowing and decreasing occurring after dam construction. The effects of daily pulses of water from peak hydropower generation and sudden sluice gate operations are investigated downstream of Hapcheon Dam in South Korea. The study reach is 45 km long from the Hapcheon re-regulation Dam to the confluence with the Nakdong River. An analysis of a time series of aerial photographs taken in 1982, 1993, and 2004 showed that the non-vegetated active channel width narrowed an average of 152 m (47% reduction since 1982). The non-vegetated active channel area also decreased an average of 6.6 $km^2$ (44% reduction since 1982) between 1982 and 2004, with most changes occurring after dam construction. The average median bed material size increased from 1.07 mm in 1983 to 5.72 mm in 2003, and the bed slope of the reach decreased from 0.000943 in 1983 to 0.000847 in 2003. The riverbed vertical degradation is approximately 2.6 m for a distance of 20 km below the re-regulation dam. It is expected from the result of the unsteady sediment transport numerical model (GSTAR-1D) steady simulations that the thalweg elevation will reach a stable condition around 2020. The model also confirms the theoretical prediction that sediment transport rates from daily pulses and flood peaks are 21 % and 15 % higher than their respective averages.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.11 no.3
• /
• pp.1-15
• /
• 2016

This study has analyzed predication capabilities leveraging multi-variate model, logistic regression model, and artificial neural network model based on financial information of medium-small sized companies list in KOSDAQ. 83 delisted companies from 2009 to 2012 and 83 normal companies, i.e. 166 firms in total were sampled for the analysis. Modelling with training data was mobilized for 100 companies inlcuding 50 delisted ones and 50 normal ones at random out of the 166 companies. The rest of samples, 66 companies, were used to verify accuracies of the models. Each model was designed by carrying out T-test with 79 financial ratios for the last 5 years and identifying 9 significant variables. T-test has shown that financial profitability variables were major variables to predict a financial risk at an early stage, and financial stability variables and financial cashflow variables were identified as additional significant variables at a later stage of insolvency. When predication capabilities of the models were compared, for training data, a logistic regression model exhibited the highest accuracy while for test data, the artificial neural networks model provided the most accurate results. There are differences between the previous researches and this study as follows. Firstly, this study considered a time-series aspect in light of the fact that failure proceeds gradually. Secondly, while previous studies constructed a multivariate discriminant model ignoring normality, this study has reviewed the regularity of the independent variables, and performed comparisons with the other models. Policy implications of this study is that the reliability for the disclosure documents is important because the simptoms of firm's fail woule be shown on financial statements according to this paper. Therefore institutional arragements for restraing moral laxity from accounting firms or its workers should be strengthened.

• Journal of the Korean earth science society
• /
• v.36 no.1
• /
• pp.90-108
• /
• 2015

The WRF-Chem model was applied to simulate the Asian dust event affecting the Korean Peninsula from 11 to 13 November 2010. GOCART dust emission schemes, RADM2 chemical mechanism, and MADE/SORGAM aerosol scheme were adopted within the WRF-Chem model to predict dust aerosol concentrations. The results in the model simulations were identified by comparing with the weather maps, satellite images, monitoring data of $PM_{10}$ concentration, and LIDAR images. The model results showed a good agreement with the long-range transport from the dust source area such as Northeastern China and Mongolia to the Korean Peninsula. Comparison of the time series of $PM_{10}$ concentration measured at Backnungdo showed that the correlation coefficient was 0.736, and the root mean square error was $192.73{\mu}g/m^3$. The spatial distribution of $PM_{10}$ concentration using the WRF-Chem model was similar to that of the $PM_{2.5}$ which were about a half of $PM_{10}$. Also, they were much alike in those of the UM-ADAM model simulated by the Korean Meteorological Administration. Meanwhile, the spatial distributions of $PM_{10}$ concentrations during the Asian dust events had relevance to those of both the wind speed of u component ($ms^{-1}$) and the PBL height (m). We performed a regressive analysis between $PM_{10}$ concentrations and two meteorological variables (u component and PBL) in the strong dust event in autumn (CASE 1, on 11 to 23 March 2010) and the weak dust event in spring (CASE 2, on 19 to 20 March 2011), respectively.

• Journal of Korea Water Resources Association
• /
• v.47 no.3
• /
• pp.297-306
• /
• 2014

In this study, EFDC (Environmental Fluid Dynamics Code) model was used to simulate the salinity distribution for sea water inflow and rainfall runoff. The flowrate was given to the boundary conditions, which can be calculated by areal-specific flowrate method from the measured flowrate of the representative outfall. The boundary condition of the water elevation can be obtained from the hourly tidal elevation. The flowrate from the outfall can be calculated using the condition of the 245 mm raifall. The simulation results showed that at Sites 1~2 and the Mangsan island (Site 4) the salinity becomes 0 ppt after the rainfall. However, the salinity is 30 ppt when there is no rainfall. Time series of the salinity changes were compared with the measured data from January 1 to December 31, 2010 at the four sites (Site 2~5) of Yongwon channel. Lower salinities are shown at the inner sites of Yongwon channel (Site 1~4) and the sites of Songjeong river (Site 7~8). The intensive investigation near the Mangsan island showed that the changes of salinity were 21.9~28.8 ppt after the rainfall of 17 mm and those of the salinity were 2.33~8.05 ppt after the cumulative rainfall of 160.5 mm. This means that the sea water circulation is blocked in Yongwon channel, and the salinity becomes lower rapidly after the heavy rain.

• Journal of KIISE:Software and Applications
• /
• v.26 no.11
• /
• pp.1261-1269
• /
• 1999

다층 퍼셉트론은 다양한 응용 분야에 성공적으로 적용되고 있는 대표적인 신경회로망 모델이다. 그러나 다층 퍼셉트론의 학습에서 나타나는 플라토에 기인한 느린 학습 속도와 지역 극소는 실제 응용문제에 적용함에 있어서 가장 큰 문제로 지적되어왔다. 이 문제를 해결하기 위해 여러 가지 다양한 학습알고리즘들이 개발되어 왔으나, 계산의 비효율성으로 인해 실제 문제에는 적용하기 힘든 예가 많은 등, 현재까지 만족할 만한 해결책은 제시되지 못하고 있다. 본 논문에서는 다층퍼셉트론의 베이시스 함수로 사용되는 시그모이드 함수를 보다 일반화된 형태로 정의하여 사용함으로써 학습에 있어서의 플라토를 완화하고, 지역극소에 빠지는 것을 줄이는 접근방법을 소개한다. 본 방법은 기존의 변형된 가중치 수정식을 사용한 학습 속도 향상의 방법들과는 다른 접근 방법을 택함으로써 기존의 방법들과 함께 사용하는 것이 가능하다는 특징을 갖고 있다. 제안하는 방법의 성능을 확인하기 위하여 간단한 패턴 인식 문제들에의 적용 실험 및 기존의 학습 속도 향상 방법을 함께 사용하여 시계열 예측 문제에 적용한 실험을 수행하였고, 그 결과로부터 제안안 방법의 효율성을 확인할 수 있었다. Abstract A multilayer perceptron is the most well-known neural network model which has been successfully applied to various fields of application. Its slow learning caused by plateau and local minima of gradient descent learning, however, have been pointed as the biggest problems in its practical use. To solve such a problem, a number of researches on learning algorithms have been conducted, but it can be said that none of satisfying solutions have been presented so far because the problems such as computational inefficiency have still been existed in these algorithms. In this paper, we propose a new learning approach to minimize the effect of plateau and reduce the possibility of getting trapped in local minima by generalizing the sigmoidal function which is used as the basis function of a multilayer perceptron. Adapting a new approach that differs from the conventional methods with revised updating equation, the proposed method can be used together with the existing methods to improve the learning performance. We conducted some experiments to test the proposed method on simple problems of pattern recognition and a problem of time series prediction, compared our results with the results of the existing methods, and confirmed that the proposed method is efficient enough to apply to the real problems.

• Journal of Korean Society of Rural Planning
• /
• v.27 no.4
• /
• pp.55-70
• /
• 2021

In this study, in the field of remote sensing, where the scope of application is rapidly expanding to fields such as land monitoring, disaster prediction, facility safety inspection, and maintenance of cultural properties, monitoring of rural space and surrounding environment using UAV is utilized. It was carried out to verify the possibility, and the following main results were derived. First, the aerial image taken with an unmanned aerial vehicle had a much higher image size and spatial resolution than the aerial image provided by the National Geographic Information Service. It was suitable for analysis due to its high accuracy. Second, the more the number of photographed photos and the more complex the terrain features, the more the point cloud included in the aerial image taken with the UAV was extracted. As the amount of point cloud increases, accurate 3D mapping is possible, For accurate 3D mapping, it is judged that a point cloud acquisition method for difficult-to-photograph parts in the air is required. Third, 3D mapping technology using point cloud is effective for monitoring rural space and rural resources because it enables observation and comparison of parts that cannot be read from general aerial images. Fourth, the digital elevation model(DEM) produced with aerial image taken with an UAV can visually express the altitude and shape of the topography of the study site, so it can be used as data to predict the effects of topographical changes due to changes in rural space. Therefore, it is possible to utilize various results using the data included in the aerial image taken by the UAV. In this study, the superiority of images acquired by UAV was verified by comparison with existing images, and the effect of 3D mapping on rural space monitoring was visually analyzed. If various types of spatial data such as GIS analysis and topographic map production are collected and utilized using data that can be acquired by unmanned aerial vehicles, it is expected to be used as basic data for rural planning to maintain and preserve the rural environment.

• Economic and Environmental Geology
• /
• v.56 no.4
• /
• pp.435-445
• /
• 2023

A delta is a depositional landform that is formed when sediment transported by a river is deposited in a relatively low-energy environment, such as a lake, sea, or a main channel. Among these, a delta formed at the confluence of rivers has a great importance in river management and research because it has a significant impact on the hydraulic and sedimentological characteristics of the river. Recently, the equilibrium state of the confluence area has been disrupted by large-scale dredging and construction of levees in the Nakdong River. However, due to the natural recovery of the river, the confluence area is returning to its pre-dredging natural state through ongoing sedimentation. The time-series data show that the confluence delta has been steadily growing since the dredging, but once it reaches a certain size, it repeats growth and retreat, and the overall size does not change significantly. In this study, we developed a model to explain the sedimentation-erosion processes in the confluence area based on the assumption that the confluence delta reaches a dynamic equilibrium. The model is based on two fundamental principles: sedimentation due to supply from the tributary and erosion due to the main channel. The erosion coefficient that represents the Nakdong River confluence areas, was obtained using data from the tributaries of the Nakdong River. Sensitivity analyses were conducted using the developed model to understand how the confluence delta responds to changes in the sediment and water discharges of the tributary and the main channel, respectively. We then used annual average discharge of the Nakdong River's tributaries to predict the dynamic equilibrium positions of the confluence deltas. Finally, we conducted a simulation experiment on the development of the Gamcheon-Nakdong River delta using recorded daily discharge. The results showed that even though it is a simple model, it accurately predicted the dynamic equilibrium positions of the confluence deltas in the Nakdong River, including the areas where the delta had not formed, and those where the delta had already formed and predicted the trend of the response of the Gamcheon-Nakdong River delta. However, the actual retreat in the Gamcheon-Nakdong River delta was not captured fully due to errors and limitations in the simplification process. The insights through this study provide basic information on the sediment supply of the Nakdong River through the confluence areas, which can be implemented as a basic model for river maintenance and management.