• Journal of Korea Port Economic Association
• /
• v.25 no.3
• /
• pp.229-250
• /
• 2009

This paper aims to implement an empirical research about the efficiency of America and Northeast Asia pots, and to suggest an effective strategy which can operate these ports more well. This study tries to apply the Data Envelopment Analysis(DEA) model to America and Northeast Asia ports. DEA is a methodology of comparing the relative efficiency of each decision making unit(DMU) by comparing it with other DMUs having similar input and output structure, and is specially very useful when a form of production function of each DMU such as a port is not known. DEA provides the extent of inefficiency of DMUs, which is practically useful information (like the efficiency score and reference sets) required to improve efficiency. This paper analyzed the relative efficiency of 35 ports in America and Northeast Asia for 3 years from 2005 to 2007 through DEA-CCR, DEA-BCC model and scale efficiency. Accordingly, this paper evaluates the efficiency of America and Northeast Asia ports, grasps the position at the present time, and suggests an advanced direction in future.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.11
• /
• pp.41-49
• /
• 2021

All application programs, including malware, call the Application Programming Interface (API) upon execution. Recently, using those characteristics, attempts to detect and classify malware based on API Call information have been actively studied. However, datasets containing API Call information require a large amount of computational cost and processing time. In addition, information that does not significantly affect the classification of malware may affect the classification accuracy of the learning model. Therefore, in this paper, we propose a method of extracting a essential feature set after reducing the dimensionality of API Call information by applying various feature selection methods. We used CICAndMal2020, a recently announced Android malware dataset, for the experiment. After extracting the essential feature set through various feature selection methods, Android malware classification was conducted using CNN (Convolutional Neural Network) and the results were analyzed. The results showed that the selected feature set or weight priority varies according to the feature selection methods. And, in the case of binary classification, malware was classified with 97% accuracy even if the feature set was reduced to 15% of the total size. In the case of multiclass classification, an average accuracy of 83% was achieved while reducing the feature set to 8% of the total size.

• Journal of Internet Computing and Services
• /
• v.20 no.4
• /
• pp.91-102
• /
• 2019

In autonomous driving systems, the ability to classify pedestrians in images captured by cameras is very important for pedestrian safety. In the past, after extracting features of pedestrians with HOG(Histogram of Oriented Gradients) or SIFT(Scale-Invariant Feature Transform), people classified them using SVM(Support Vector Machine). However, extracting pedestrian characteristics in such a handcrafted manner has many limitations. Therefore, this paper proposes a method to classify pedestrians reliably and effectively using CNN's(Convolutional Neural Network) deep features and transfer learning. We have experimented with both the fixed feature extractor and the fine-tuning methods, which are two representative transfer learning techniques. Particularly, in the fine-tuning method, we have added a new scheme, called M-Fine(Modified Fine-tuning), which divideslayers into transferred parts and non-transferred parts in three different sizes, and adjusts weights only for layers belonging to non-transferred parts. Experiments on INRIA Person data set with five CNN models(VGGNet, DenseNet, Inception V3, Xception, and MobileNet) showed that CNN's deep features perform better than handcrafted features such as HOG and SIFT, and that the accuracy of Xception (threshold = 0.5) isthe highest at 99.61%. MobileNet, which achieved similar performance to Xception and learned 80% fewer parameters, was the best in terms of efficiency. Among the three transfer learning schemes tested above, the performance of the fine-tuning method was the best. The performance of the M-Fine method was comparable to or slightly lower than that of the fine-tuningmethod, but higher than that of the fixed feature extractor method.

• Journal of Korea Water Resources Association
• /
• v.52 no.3
• /
• pp.187-194
• /
• 2019

Intake ratio from river constitutes about 31% (8/26) that beings to "water stress country" as "Medium ~ High" with China, India, Italy, South Africa, etc. Therefore, the present study on a dam that is the most effective and direct for securing water resources has been performed. First of all, climate change scenarios were investigated and analyzed. RCP 4.5 and 8.5 with 12.5 km grid resolution presented in the IPCC (Intergovernmental Panel on Climate Change) 5th Assessment Report (AR5) were applied to study watershed using SWAT (Soil and Water Assessment Tool) and HEC-ResSim models that carried out co-operation. Based on the results of dam simulation, the reduction effects of floods and droughts were quantitatively presented. The procedures of dam projects of the USA, Japan and Korea were investigated. As a result, there are no estimating quantitative criteria, calculating methods or formulas. In the present study, therefore, indexes for selecting suitable dam site through literature investigation and analyzing dam watersheds were determined, Expert questionnaire for various indexes were performed. Based on the above mentioned investigation and expert questionnaire, a methodology assigning weight using AHP method were proposed. The function of suitable dam (FSDS) site was calibrated and verified for four medium-sized watersheds. Finally, automated model for suitable dam site was developed using FSDS and 'Model builder' of GIS tool.

• Journal of Korean Society of Forest Science
• /
• v.100 no.4
• /
• pp.664-671
• /
• 2011

Random Walk Model can predict the sediment areas of debris flow but it must be extracted three parameters fitted topographical environment. This study developed the method to extract the optimal values of three parameters - Once flowing volume, Stopping slope and Gravity weight - for Random Walk Model. And the extracted parameters were validated by aerial photographs of the debris flowed area. To extract the optimal parameters was randomly performed, limiting the range values of three parameters and developing an accuracy decision method that is called the rate of concordance. The set of the optimal parameters was decided on highest the rate of concordance and a consistency. As a result, the optimal parameters in Bonghwa county were showed that the once flowing volume is $1.0m^3$, the stopping slope is $4.2^{\circ}$ and the gravity weight is 2 when the rate of concordance is -0.2. The validating result of the optimal parameters showed closely that the rate of concordance is average -0.2.

• Journal of the Korea Convergence Society
• /
• v.12 no.3
• /
• pp.9-15
• /
• 2021

With the recent development of 5G and artificial intelligence technologies, it is interested in AIOT technology to collect, process, and analyze information in cloud edge environments. AIIoT technology is being applied to various smart environments, but research is needed to perform fast response processing through accurate analysis of collected information. In this paper, we propose a technique to minimize bandwidth and processing time by blocking the connection processing between AIOT information through fast processing and accurate analysis/forecasting of information collected in the smart environment. The proposed technique generates seeds for data indexes on AIOT devices by multipointing information collected by blockchain, and blocks them along with collection information to deliver them to the data center. At this time, we deploy Deep Neural Network (DNN) models between cloud and AIOT devices to reduce network overhead. Furthermore, server/data centers have improved the accuracy of inaccurate AIIoT information through the analysis and predicted results delivered to minimize latency. Furthermore, the proposed technique minimizes data latency by allowing it to be partitioned into a layered multilayer network because it groups it into blockchain by applying weights to AIOT information.

• Journal of Korea Port Economic Association
• /
• v.37 no.4
• /
• pp.103-126
• /
• 2021

The purpose of this study is described in detail as follows. First, I would like to define what digital transformation is in the maritime transport sector. Second, it is intended to derive success factors for digital transformation in the maritime transportation field by examining various preceding studies related to digital transformation. Finally, in order to derive priorities for the derived success factors, an AHP analysis model is built and an expert survey is conducted for practical experts in the maritime transportation field. Based on the survey results, we would like to provide guidelines on what factors should be considered first among the success factors of digital transformation in the maritime transportation sector. In this study, in order to derive the priority of success factors for digital transformation in the maritime transportation field, the hierarchical structure was divided into four high-level evaluation items(strategic factors, organizational culture and human factors, technology factors, and environmental factors) and 21 sub-evaluation items. A relative evaluation method of weighting items among AHP(Analytic Hierarchy Process) was applied. AHP analysis of 24 questionnaires with a consistency ratio of 0.1 or less in order to increase the accuracy of information among questionnaires collected through maritime transportation related university professors, research groups, shipping companies, container terminals, and experts engaged in shipping related IT companies was carried out. As a result of the analysis, the priority of the first-tier factors for the success factors of digital transformation in the maritime transport sector was shown in the order of strategic factors, organizational culture and human factors, technology factors, and environmental factors. In addition, when looking at the priorities of 21 detailed items, it was found that the development of new business models, the creation of an active future digital strategy, and the leadership of the chief digital officer were high.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.1
• /
• pp.51-58
• /
• 2023

In this paper, we propose an organ segmentation technique for the automatic extraction of medical diagnostic indicators from X-ray images. In order to calculate diagnostic indicators of heart disease and spinal disease such as VHS(vertebral heart scale) and Cobb angle, it is necessary to accurately segment the thoracic spine, carina, and heart in a chest X-ray image. A deep neural network model in which the high-resolution representation of the image for each layer and the structure converted into a low-resolution feature map are connected in parallel was adopted. This structure enables the relative position information in the image to be effectively reflected in the segmentation process. It is shown that learning performance can be improved by combining the OCR module, in which pixel information and object information are mutually interacted in a multi-step process, and the channel attention module, which allows each channel of the network to be reflected as different weight values. In addition, a method of augmenting learning data is presented in order to provide robust performance against changes in the position, shape, and size of the subject in the X-ray image. The effectiveness of the proposed theory was evaluated through an experiment using 145 human chest X-ray images and 118 animal X-ray images.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.1A
• /
• pp.1-9
• /
• 2006

An analysis of the annual frequency of collapse(AF) is performed for each bridge pier exposed to ship collision. From this analysis, the impact lateral resistance can be determined for each pier. The bridge pier impact resistance is selected using a probability-based analysis procedure in which the predicted annual frequency of bridge collapse, AF, from the ship collision risk assessment is compared to an acceptance criterion. The analysis procedure is an iterative process in which a trial impact resistance is selected for a bridge component and a computed AF is compared to the acceptance criterion, and revisions to the analysis variables are made as necessary to achieve compliance. The distribution of the AF acceptance criterion among the exposed piers is generally based on the designer's judgment. In this study, the acceptance criterion is allocated to each pier using allocation weights based on the previous predictions. To determine the design impact lateral resistance of bridge components such pylon and pier, the numerical analysis is performed iteratively with the analysis variable of impact resistance ratio of pylon to pier. The design impact lateral resistance can vary greatly among the components of the same bridge, depending upon the waterway geometry, available water depth, bridge geometry, and vessel traffic characteristics. More researches on the allocation model of AF and the determination of impact resistance are required.

• 한국노년학
• /
• v.29 no.2
• /
• pp.579-591
• /
• 2009

The purpose of this study is what are the factors affecting decision making for selection of social welfare facility, esp., elderly care facility. The resident's selection for social welfare facility are relatively new terms in the Korean social work. To explore the factors, we employed an empirical study. The collected data was analyzed by using SEM(Structural equation modeling). As the results, the standardized regression coefficient of the hypothesis that the service quality will affect the service satisfaction is equal to 0.555(t=6.723, p<0.01) and the coefficient that the service satisfaction will affect the decision making is equal to 0.766. The absolute fitness of the SEM model shows χ² = 580.151(d.f.=317, p-value=0.000) and RMSEA=0.063, so that the model is absolutely fit. Since CFI=0.915, the model is comparatively fit. Conclusively speaking, the hypothesis that service quality affects to service satisfaction was supported. The hypothesis that service satisfaction affects to decision making was also supported. By accounting for the results of the analysis, this study verified the service quality leading to service satisfaction is an important factor for resident to select a residential facility.