• Journal of Arbitration Studies
• /
• v.30 no.4
• /
• pp.121-138
• /
• 2020

The O2O services in the healthcare sector have only been in full swing for about three years, and unlike existing O2O consumer goods, the scale and scope of the dispute are more complicated due to restrictions on medical treatment. In this study, O2O service platform operators and medical institutions' roles and responsibilities were redefined as a countermeasure for resolving disputes in healthcare O2O services and the laws for changing the transaction environment. A change in institutional mechanisms was proposed. This study looked at the types of consumer disputes related to healthcare O2O services as insufficient information problems, problems in the course of medical service implementation, problems with immunity provisions for platform operators, cancellations, and non-compliance with refunds. All the information generated during transactions in the healthcare sector was extensive in scale and included the most sensitive information among personal information, stressing the importance of ensuring security. The area that started in the O2O range before the medical institution visit also proposed a plan to establish a system for the delivery of proven information as a pre-medical person. The scale and growth will grow faster, given that consumers can experience the information they want anytime, anywhere they want. However, the platform broker's role, a link player, will become more important because consumers who use the service will have their first meeting with non-face-to-face product providers. On the other hand, service providers may have side effects of misleading consumers by providing false information or misleading consumers through exaggerated advertisements. The O2O service market is expected to expand beyond distribution and dining out to the entire industry. However, since it is challenging to check accurate statistics on the detailed market, various disputes and consumer protection measures will be required for each detailed market, and comprehensive leading solutions will be essential in the future.

• Geomechanics and Engineering
• /
• v.28 no.5
• /
• pp.437-454
• /
• 2022

The hydro-mechanical (HM) two-way sequential coupling in the TOUGH2-FLAC3D linking algorithm is validated completely and successfully in both M to H and H to M directions, which are initiated by mechanical surface loading for geomechanical validation and hydrological groundwater pumping for hydrogeological validation, respectively. For such complete and successful validation, a TOUGH2-FLAC3D linked numerical model is developed first by adopting the TOUGH2-FLAC3D linking algorithm, which uses the two-way (fixed-stress split) sequential coupling scheme and the implicit backward time stepping method. Two geomechanical and two hydrogeological validation problems are then simulated using the linked numerical model together with basic validation strategies and prerequisites. The second geomechanical and second hydrogeological validation problems are also associated with the Mandel effect and the Noordbergum and Rhade effects, respectively, which are three phenomenally well-known but numerically challenging HM effects. Finally, sequentially coupled numerical solutions are compared with either analytical solutions (verification) or fully coupled numerical solutions (benchmarking). In all the four validation problems, they show almost perfect to extremely or very good agreement. In addition, the second geomechanical validation problem clearly displays the Mandel effect and suggests a proper or minimum geometrical ratio of the height to the width for the rectangular domain to maximize agreement between the numerical and analytical solutions. In the meantime, the second hydrogeological validation problem clearly displays the Noordbergum and Rhade effects and implies that the HM two-way sequential coupling scheme used in the linked numerical model is as rigorous as the HM two-way full coupling scheme used in a fully coupled numerical model.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.6 s.312
• /
• pp.34-42
• /
• 2006

As art analytic method to uncover interesting patterns hidden under a large volume of data, data mining research has been actively done so far in various fields. However, current state-of-the-arts in data mining research have several challenging problems such as being too ad-hoc. The existing techniques are mostly the ones designed for individual problems, so there is no unifying theory applicable for more general data mining problems. In this paper, we address the problem of classification, which is one of significant data mining tasks. Specifically, our objective is to evaluate radial basis function (RBF) model for classification tasks and investigate its usefulness. For evaluation, we analyze the popular Monk's problems which are well-known datasets in data mining research. First, we develop RBF models by using the representational capacity based learning algorithm, and then perform a comparative assessment of the results with other models generated by the existing techniques. Through a variety of experiments, it is empirically shown that the RBF model has not only the superior performance on the Monk's problems but also its modeling process can be controlled in a systematic way, so the RBF model with RC-based algorithm might be a good candidate to handle the current ad-hoc problem.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.4
• /
• pp.515-524
• /
• 2022

To effectively collect, manage, and share the maritime traffic information, it is necessary to identify the technology trends concerning this particular information and analyze its current status and problems. Therefore, this study observes the domestic and foreign technology trends involving maritime traffic information while analyzing and summarizing the current status and problems in collecting, managing, and sharing it. According to the data analysis, the problems in the collecting stage are difficulties in collecting visual information from long-distance radars, CCTVs, and cameras in areas outside the LTE network coverage. Notably, this explains the challenges in detecting smuggling ships entering the territorial waters through the exclusive economic zone (EEZ) in the early stage. The problems in the management stage include difficult reductions and expansions of maritime traffic information caused by the lack of flexibility in storage spaces mostly constructed by the maritime transportation system. Additionally, it is challenging to deal with system failure with system redundancy and backup as a countermeasure. Furthermore, the problems in the sharing stage show that it is difficult to share information with external operating organizations since the internal network is mainly used to share maritime transportation information. If at all through the government cloud via platforms such as LRIT and SASS, it often fails to effectively provide various S/W applications that help use maritime big data. Therefore, it is suggested that collecting equipment such as unmanned aerial vehicles and satellites should be constructed to expand collecting areas in the collecting stage. In the management and sharing stages, the introduction and construction of private clouds are suggested, considering the operational administration and information disclosure of each maritime transportation system. Through these efforts, an enhancement of the expertise and security of clouds is expected.

• Journal of Communications and Networks
• /
• v.18 no.5
• /
• pp.744-761
• /
• 2016

For about the past decade and a half research efforts into cognitive radio networks (CRNs) have increased dramatically. This is because CRN is recognized as a technology that has the potential to squeeze the most out of the existing spectrum and hence virtually increase the effective capacity of a wireless communication system. The resulting increased capacity is still a limited resource and its optimal allocation is a critical requirement in order to realize its full benefits. Allocating these additional resources to the secondary users (SUs) in a CRN is an extremely challenging task and integer programming based optimization tools have to be employed to achieve the goals which include, among several aspects, increasing SUs throughput without interfering with the activities of primary users. The theory of the optimization tools that can be used for resource allocations (RA) in CRN have been well established in the literature; convex programming is one of them, in fact the major one. However when it comes to application and implementation, it is noticed that the practical problems do not fit exactly into the format of well established tools and researchers have to apply approximations of different forms to assist in the process. In this survey paper, the optimization tools that have been applied to RA in CRNs are reviewed. In some instances the limitations of techniques used are pointed out and creative tools developed by researchers to solve the problems are identified. Some ideas of tools to be considered by researchers are suggested, and direction for future research in this area in order to improve on the existing tools are presented.

• Korean Journal of Health Education and Promotion
• /
• v.3 no.1
• /
• pp.63-73
• /
• 1985

Ratio of physically and mentally handicapped people among the total population is about 3∼4%. Regardless of the extent of seriousness of handicaps, they have to sustain various difficulties in their life time and pose challenging problems to their homes, community and nation. In the spectrum of their problems, measures such as treatment, rehabilitation and welfare work all occupied important part but most significant measure is the prevention. Such handicaps items from health care of maternal and child, mostly causes throughout the process of pregnancy, delivery and upbringing. And then this study is focused on the pregnant history of the mothers with handicapped children. The objective of this study is to identify the basic factors which can create a handicapped child during pregnancy and is to prevent the birth of such a child. For this study, 301 mothers who have handicapped children were selected, now receiving various training at 5 rehabilitation and educational facilities. And questionnaires distributed to them during the period of Oct, 1983 to Dec, 14. The collected responses have been analized in terms of significant numbers, percentage and metic average. Their related results are as follows: 1. The sexual ratio between the male and female was 58.5% compare 41.5%. The average age of the handicapped children was 10.1 years old and that of mothers at the time of the children's birth was 28.9 old years. They were born as 2.14th in their family, the ratio of antenatal care for pregnant mothers was 36.2% and 22.9% were abnormally delivered. 37.9% out of the total were born from mothers with ages above 35 years and as their first baby. 2. The time of confirming the handicapped nature of children was during babyhood by 97.1% and discovery of these misfortunes were made by chance. As for causes of those-handicapped status 20.9% disease from complication, 15.3% from difficulties experience at the time of delivery, 11.3% from heredity and 10% were results of drug abuse during pregnancy, and 49.8% of the handicapped children were mentally retarded.

• Archives of Reconstructive Microsurgery
• /
• v.12 no.1
• /
• pp.19-29
• /
• 2003

Soft tissue reconstruction of dorsum of the foot and ankle has long presented challenging problems for the reconstructive surgeon. Limitations of available local tissue, the need for specialized tissue, and donor site morbidity restrict the options. In an effort to solve these difficult problems, we have begun to use adipofascial flap based on the perforating branch of the peroneal artery. We present our early experience of 5 patients treated with this flap. Our patients ranged from 6 to 26 years in age and included 3 males and 2 females. The etiologies of the wounds were secondary to traffic accident, and crushing injury. The flaps had reverse flow in all patients. The flap and the adjoining raw area were covered with a full-thickness skin graft, while the donor site at the lateral aspect of the leg was closed primarily without grafting. The skin graft was taken from the inguinal area, which was closed primarily. Compared with other flap, this adipofascial flap is thinner, producing less bulkiness to the recipient site and minor aesthetic sequelae to the donor site. In our opinion, this flap is versatile, effective, and an addition to the armamentarium of the reconstructive surgeon for coverage of difficult wounds of the foot and ankle.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.27 no.2
• /
• pp.213-222
• /
• 2017

Online advertisement has many benefits comparing to offline advertisement but it also has many challenging problems by online ad abuses. Advertisement injection (Ad injection) is one of the threats that surreptitiously inserts advertisements without a permission of site owners. Users are exposed to additional ads and redundant web traffic by injected ads can cause a service quality problem. Moreover, advertisers can have economic loss when injected ads are different from original ones. Although ad injection leads to these problems it has not been fully studied yet. A few ad injection researches are done by online advertising providers such as Google. In this paper, we analyze ad injection activities to Korean major portal, Naver. We classify 6 types of ad injections and describe their characteristics by analyzing 27 downloaders and 199 installed programs.

• Journal of KIISE
• /
• v.44 no.7
• /
• pp.680-685
• /
• 2017

The emergence of online media and their data has enabled data-driven methods to solve challenging and complex tasks such as rumor classification problems. Recently, deep learning based models have been shown as one of the fastest and the most accurate algorithms to solve such problems. These new models, however, either rely on complete data or several days-worth of data, limiting their applicability in real time. In this study, we go beyond this limit and test the possibility of super early rumor detection via recurrent neural networks (RNNs). Our model takes in social media streams as time series input, along with basic meta-information about the rumongers including the follower count and the psycholinguistic traits of rumor content itself. Based on analyzing millions of social media posts on 498 real rumors and 494 non-rumor events, our RNN-based model detected rumors with only 30 initial posts (i.e., within a few hours of rumor circulation) with remarkable F1 score of 0.74. This finding widens the scope of new possibilities for building a fast and efficient rumor detection system.

• Journal of Korea Water Resources Association
• /
• v.31 no.5
• /
• pp.599-612
• /
• 1998

Numerical analysis of convection-dominated transport problems are challenging because of dual characteristics of the governing equation. In the finite element method, a strategy is to modify the test function to weight more in the upwind direction. This is called as the Petrov-Galerkin method. In this paper, both N+1 and N+2 Petrov-Galerkin methods are applied to transport problems at high grid Peclet number. Frequency fitting algorithm is used to obtain optimal levels of N+2 upwinding, and the results are discussed. Also, a new Petrov-Galerkin method, named as "Optimal Test Function Petrov-Galerkin Method," is proposed in this paper. The test function of this numerical method changes its shape depending upon relative strength of the convection to the diffusion. A numerical experiment is carried out to demonstrate the performance of the proposed method.