• Proceedings of the Korea Water Resources Association Conference
• /
• 2019.05a
• /
• pp.157-157
• /
• 2019

This study derived the Flood-Inducing-Rainfall (FIR) and the Flood-Inducing-Runoff (FIRO) from the radar-gage composite data to be used as the basis of the flood warning initiation for the urban area of Seoul. For this, we derived the rainfall depth-duration relationship for the 261 flood events at 239 watersheds during the years 2010 and 2011 based on the 10-minute 1km-1km radar-gauge composite rainfall field. The relationship was further refined by the discrete ranges of the proportion of the flooded area in the watershed (FP) and the coefficient variation of the rainfall time series (CV). Then, the slope of the straight line that contains all data points in the depth-duration relationship plot was determined as the FIR for the specified range of the FP and the CV. Similar methodology was applied to derive the FIRO, which used the runoff depths that were estimated using the NRCS Curve Number method. We found that FIR and FIRO vary at the range of 37mm/hr-63mm/hr and the range of 10mm/hr-42mm/hr, respectively. The large variability was well explained by the FP and the CV: As the FP increases, FIR and FIRO increased too, suggesting that the greater rainfall causes larger flooded area; as the rainfall CV increases, FIR and FIRO decreased, which suggests that the temporally concentrated rainfall requires less total of rainfall to cause the flood in the area. We verified our result against the 21 flood events that occurred for the period of 2012 through 2015 for the same study area. When the 5 percent of the flooded area was tolerated, the ratio of hit-and-miss of the warning system based on the rainfall was 44.2 percent and 9.5 percent, respectively. The ratio of hit-and-miss of the warning system based on the runoff was 67 percent and 4.7 percent, respectively. Lastly, we showed the importance of considering the radar-gauge composite rainfall data as well as rainfall and runoff temporal variability in flood warning system by comparing our results to the ones based on the gauge-only or radar-only rainfall data and to the one that does not account for the temporal variability.

• Journal of the Korea Convergence Society
• /
• v.10 no.4
• /
• pp.1-6
• /
• 2019

Neural networks have been used to predict the direction of stock index movement from past data. The conventional research that predicts the upward or downward movement of the stock index predicts a rise or fall even with small changes in the index. It is highly likely that losses will occur when trading ETFs by use of the prediction. In this paper, a neural network model that predicts the movement direction of the daily KOrea composite Stock Price Index (KOSPI) to reduce ETF trading losses and earn more than a certain amount per trading is presented. The proposed model has outputs that represent rising (change rate in index ${\geq}{\alpha}$), falling (change rate ${\leq}-{\alpha}$) and neutral ($-{\alpha}$ change rate < ${\alpha}$). If the forecast is rising, buy the Leveraged Exchange Traded Fund (ETF); if it is falling, buy the inverse ETF. The hit ratio (HR) of PNN1 implemented in this paper is 0.720 and 0.616 in the learning and the evaluation respectively. ETF trading yields a yield of 8.386 to 16.324 %. The proposed models show the better ETF trading success rate and yield than the neural network models predicting KOSPI.

• Journal of Korean Society of Forest Science
• /
• v.100 no.3
• /
• pp.374-381
• /
• 2011

The purpose of this study is to examine the possibility of early detection of Pine Wilt Disease (PWD) using NDVI (Normalized Difference Vegetation Index) from ADC (Agricultural Digital Camera) imageries. The PWD induces the different patterns of reduction of NDVI between healthy trees and infected trees, due to the withered leaves on the infected trees. Based on these phenomena, the DI showing the NDVI variations of trees by time series was employed to detect the infected trees. To find out the differences of DI level between normal and infected trees, DIs of trees from May to August in 2007 were calculated and these were analyzed with GLM (General Linear Models) in SAS 9.2. As a result, the difference of DI between in June and August shows the most significant level (0.0001). The discriminant analysis was performed between normal and infected trees, using the DI of June and August. As the result, hit ratio of trees and the accuracy of grouping with Jack-knife method were shown 71.9% and 73.5%, respectively. These results showed that the DI is effective to detect the trees infected by the PWD and it is useful to prevent the PWD.

• Journal of Service Research and Studies
• /
• v.13 no.2
• /
• pp.99-113
• /
• 2023

In a customer-based industry, customer retention is the competitiveness of a company, and improving customer retention improves the competitiveness of the company. Therefore, accurate prediction and management of potential dormant customers is paramount to increasing the competitiveness of the enterprise. In particular, there are numerous competitors in the domestic card industry, and the government is introducing an automatic closing system for dormant card management. As a result of these social changes, the card industry must focus on better predicting and managing potential dormant cards, and better predicting dormant customers is emerging as an important challenge. In this study, the Recurrent Neural Network (RNN) methodology was used to predict potential dormant customers in the card industry, and in particular, Long-Short Term Memory (LSTM) was used to efficiently learn data for a long time. In addition, to redefine the variables needed to predict dormant customers in the card industry, Unified Theory of Technology (UTAUT), an integrated technology acceptance theory, was applied to redefine and group the variables used in the model. As a result, stable model accuracy and F-1 score were obtained, and Hit-Ratio proved that models using LSTM can produce stable results compared to other algorithms. It was also found that there was no moderating effect of demographic information that could occur in UTAUT, which was pointed out in previous studies. Therefore, among variable selection models using UTAUT, dormant customer prediction models using LSTM are proven to have non-biased stable results. This study revealed that there may be academic contributions to the prediction of dormant customers using LSTM algorithms that can learn well from previously untried time series data. In addition, it is a good example to show that it is possible to respond to customers who are preemptively dormant in terms of customer management because it is predicted at a time difference with the actual dormant capture, and it is expected to contribute greatly to the industry.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.11 no.1
• /
• pp.73-80
• /
• 2016

The Car License Plate Recognition(: CLPR) is required in searching the hit-and-run car, measuring the traffic density, investigating the traffic accidents as well as in pursuing vehicle crimes according to the increasing in number of vehicles. The captured images on the real environment of the CLPR is contaminated not only by snow and rain, illumination changes, but also by the geometrical distortion due to the pose changes between camera and car at the moment of image capturing. We propose homographic transformation and intensity histogram of vertical image projection so as to transform the distorted input to the original image and cluster the character and number, respectively. Especially, in this paper, the Multilayer Perceptron Algorithm(: MLP) in the CLPR is used to not only recognize the charcters and car license plate, but also determine the optimized ratio among the number of input, hidden and output layers by the real experimental result.

• Structural Engineering and Mechanics
• /
• v.82 no.2
• /
• pp.173-189
• /
• 2022

This study presents an experimental and numerically study about the effects of fiber reinforcement ratio on the behavior of concrete-filled steel tubes (CFST) under dynamic impact loading. In literature have examined the behavior of GFRP and FRP wrapped strengthened CFST elements impact loads. However, since the direction of potential impact force isn't too exact, there is always the probability of not being matched the impact force of the area where the reinforced. Therefore, instead of the fiber textile wrapping method which strengthens only a particular area of CFST element, we used fiber-added concrete-filled elements which allow strengthening the whole element. Thus, the effect of fiber-addition in concrete on the behavior of CFST elements under impact loads was examined. To do so, six simply supported CFST beams were constructed with none fiber, 2% fiber and 10% fiber reinforcement ratio on the concrete part of the CFST beam. CFST beams were examined under two different impact loads (75 kg and 225 kg). The impactors hit the beam from a 2000 mm free fall during the experimental study. Numerical models of the specimens were created using ABAQUS finite element software and validated with experimental data. The obtained results such as; mid-span displacement, acceleration, failure modes and energies from experimental and numerical studies were compared and discussed. Furthermore, the Von Misses stress distribution of the CFST beams with different ratio of fiber reinforcements were investigated numerically. To sum up, there is an optimum amount limit of the fiber reinforcement on CFST beams. Up to this limit, the fiber reinforcement increases the structural performances of the beam, beyond that limit the fiber reinforcement decreases the performances of the CFST beam under transverse impact loadings.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.12
• /
• pp.1815-1827
• /
• 1993

In this paper, we evaluate the performance of a Futurebus based multiprocessor system with MESI cache coherence protocol for four bus transaction types. Graphical symbols and compiler of SLAM II are used in modeling and simulation. A steady-state probability of each state for MESI protocol is computed by a Markov chain. The probability of each state is used as an input value for a correct simulation. Processor utilization, memory utilization, bus utilization, and the waiting time for bus arbitration are measured in terms of the number of processors, the hit ratio of cache memory, the probability of internal operation, and bus bandwidth.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.3 no.1
• /
• pp.119-133
• /
• 2009

The purpose of this research is to accurately estimate the location of a device using the received signal strength indicator (RSSI) of IEEE 802.11 WLAN for location tracking in indoor environments. For the location estimation method, we adopted the calibration model. By applying the Adaptive Zone Based K-NNSS (AZ-NNSS) algorithm, which considers the velocity of devices, this paper presents a 9% improvement of accuracy compared to the existing K-NNSS-based research, with 37% of the K-NNSS computation load. The accuracy is further enhanced by using a Kalman filter; the improvement was about 24%. This research also shows the level of accuracy that can be achieved by replacing a subset of the calibration data with values computed by a numerical equation, and suggests a reasonable number of calibration points. In addition, we use both the mean error distance (MED) and hit ratio to evaluate the accuracy of location estimation, while avoiding a biased comparison.

• ETRI Journal
• /
• v.36 no.5
• /
• pp.772-782
• /
• 2014

Two-microphone binary mask speech enhancement (2mBMSE) has been of particular interest in recent literature and has shown promising results. Current 2mBMSE systems rely on spatial cues of speech and noise sources. Although these cues are helpful for directional noise sources, they lose their efficiency in diffuse noise fields. We propose a new system that is effective in both directional and diffuse noise conditions. The system exploits two features. The first determines whether a given time-frequency (T-F) unit of the input spectrum is dominated by a diffuse or directional source. A diffuse signal is certainly a noise signal, but a directional signal could correspond to a noise or speech source. The second feature discriminates between T-F units dominated by speech or directional noise signals. Speech enhancement is performed using a binary mask, calculated based on the proposed features. In both directional and diffuse noise fields, the proposed system segregates speech T-F units with hit rates above 85%. It outperforms previous solutions in terms of signal-to-noise ratio and perceptual evaluation of speech quality improvement, especially in diffuse noise conditions.

• Journal of Korean Port Research
• /
• v.13 no.1
• /
• pp.101-112
• /
• 1999

Traffic accidents increase with the increase of the vehicles in operation on the street. Especially big traffic accidents composed of over 3 killed or 20 injured accidents with the property damage become one of the serious problems to be solved in most of the cities. The purpose of this study is to build the discrimination model on big traffic accidents using the Quantification II theory for establishing the countermeasures to reduce the big traffic accidents. The results are summarized as follows. 1)The existing traffic accident related model could not explain the phenomena of the current traffic accident appropriately. 2) Based on the big traffic accident types vehicle-vehicle, vehicle-alone, vehicle-pedestrian and vehicle-train accident rates 73%, 20.5% 5.6% and two cases respectively. Based on the law violation types safety driving non-fulfillment center line invasion excess speed and signal disobedience were 48.8%, 38.1% 2.8% and 2.8% respectively. 3) Based on the law violation types major factors in big traffic accidents were road and environment, human, and vehicle in order. Those factors were vehicle, road and environment, and human in order based on types of injured driver’s death. 4) Based on the law violation types total hitting and correlation rates of the model were 53.57% and 0.97853. Based on the types of injured driver’s death total hitting and correlation rates of the model were also 71.4% and 0.59583.