• Journal of the Korea Society of Computer and Information
• /
• v.23 no.12
• /
• pp.81-87
• /
• 2018

Cloud computing is a computing environment in which users borrow as many IT resources as they need to, and use them over the network at any point in time. This is the concept of leasing and using as many IT resources as needed to lower IT resource usage costs and increase efficiency. Recently, cloud computing is emerging to provide stable service and volume of data along with major technological developments such as the Internet of Things, artificial intelligence and big data. However, for a more secure cloud environment, the importance of perimeter security such as shared resources and resulting secure data storage and access control is growing. This paper analyzes security threats in cloud computing environments and proposes a security architecture for effective response.

• Journal of computational fluids engineering
• /
• v.6 no.3
• /
• pp.10-18
• /
• 2001

Genetic Algorithm(GA) is applied to aerodynamic shape optimization and demonstrated its merits in global searching ability and the independency of differentiability. However, applications of GA are limited due to slow convergence rate, premature termination, and high computing costs. The present aerodynamic designs such as wing shape optimizations using GA have seldom been applied because of high computing costs. This paper has two objects; improvement of the efficiency of GA and application of GA into aerodynamic shape optimization for 2D and 3D wings. The study indicates that GA can be applied to aerodynamic design and its performance is comparable to traditional design methods.

• Journal of Electrical Engineering and Technology
• /
• v.5 no.3
• /
• pp.363-370
• /
• 2010

Load forecasting has always been essential to the operation and planning of power systems in deregulated electricity markets. Various methods have been proposed for load forecasting, and the neural network is one of the most widely accepted and used techniques. However, to obtain more accurate results, more information is needed as input variables, resulting in huge computational costs in the learning process. In this paper, to reduce training time in multi-layer perceptron-based short-term load forecasting, a graphics processing unit (GPU)-based computing method is introduced. The proposed approach is tested using the Korea electricity market historical demand data set. Results show that GPU-based computing greatly reduces computational costs.

• Journal of Preventive Medicine and Public Health
• /
• v.34 no.3
• /
• pp.183-190
• /
• 2001

Objective : To estimate the annual economic costs attributable to cigarette smoking in Korea. Methods : The costs were classified as being direct medical and non-medical costs, indirect costs and others. We focused on those costs related that are incurred in the treatment of selected diseases (cardiovascular diseases, respiratory diseases, and cancers), which have been proven to be caused by smoking. In addition to the basic costs of treatment, the additional amount of costs occurred due to smoking was obtained by computing the population attributable risk (PAR%) caused by smoking. To compute the PAR%, relative risks of smoking to the number of outpatient visits, hospitalizations, and the death were estimated using the Cox proportional hazard model, respectively. Our major data source was the 'Korea Medical Insurance Corporation (KMIC) cohort study,' which was composed of a total of 115,682 male and 67,932 female beneficiaries who had complete records of their smoking histories in the year of 1992. Results : The annual costs that could be attributable to smoking were estimated to be in the range of 2,847,500 million Won to 3,959,100 million Won. The maximum estimate of 3,959,100 million Won includes 233,100 million Won for medical costs, 5,100 million Won for transportation costs, 27,600 million Won for care giver's economic costs, 69,100 million Won in productivity loss, 3,435,000 million Won lost because of premature death, 172,100 million Won in costs resulting from passive smoke inhalation and 17,100 million Won for costs that resulted from fires that were caused by careless smoking. Conclusion : Our study confirms that the magnitude of the economic burden of smoking to Korean society is substantial. Therefore, this study provides strong evidence that there is a strong need for a national policy of tobacco control in Korea.

• Journal of Korea Multimedia Society
• /
• v.1 no.1
• /
• pp.120-129
• /
• 1998

The mobile computing is a new network environment that support s wireless network connection seamless while the portable computer is moving. This environment involves the problem of re-establishing of new network connection. The mobile computing should provide multicasting service, because multicasting is widely-used application service which delivers messages to the members of the groups. In this paper, we proposed 3-tiered mobile multicasting protocol that transfers multicast packets efficiently and reliablely. Also, we analyzed the performance of 3-tiered mobile multicasting protocol comparing with other mechanism. Our protocol is superior to other mechanism in the aspects of the handoff costs and packet delivery costs.

• International Journal of Computer Science & Network Security
• /
• v.22 no.3
• /
• pp.29-36
• /
• 2022

In deep learning classification tasks, most models frequently assume that all labels are available for the training datasets. As such strategies to learn new concepts from unlabeled datasets are scarce. In fingerprint classification tasks, most of the fingerprint datasets are labelled using the subject/individual and fingerprint datasets labelled with finger type classes are scarce. In this paper, authors have developed approaches of classifying fingerprint images using the majorly known fingerprint classes. Our study provides a flexible method to learn new classes of fingerprints. Our classifier model combines both the clustering technique and use of deep learning to cluster and hence label the fingerprint images into appropriate classes. The K means clustering strategy explores the label uncertainty and high-density regions from unlabeled data to be clustered. Using similarity index, five clusters are created. Deep learning is then used to train a model using a publicly known fingerprint dataset with known finger class types. A prediction technique is then employed to predict the classes of the clusters from the trained model. Our proposed model is better and has less computational costs in learning new classes and hence significantly saving on labelling costs of fingerprint images.

• International Journal of Advanced Culture Technology
• /
• v.6 no.4
• /
• pp.226-232
• /
• 2018

The rapid pace of growth in internet usage and rich mobile applications and with the advantage of incredible usage of internet enabled mobile devices the Green Mobile Crowd Computing will be the suitable area to research combining with cloud services architecture. Our proposed Framework will deploy the eHealth among various health care sectors and pave a way to create a Green Mobile Application to provide a better and secured way to access the Products/ Information/ Knowledge, eHealth services, experts / doctors globally. This green mobile crowd computing and cloud architecture for healthcare information systems are expected to lower costs, improve efficiency and reduce error by also providing better consumer care and service with great transparency to the patient universally in the field of medical health information technology. Here we introduced novel architecture to use of cloud services with crowd sourcing.

• Convergence Security Journal
• /
• v.12 no.3
• /
• pp.59-65
• /
• 2012

Clouding computing which is developing newly as IT and network technology develops become changed to internet and service environment of company. Especially, it can lend IT resource at low costs and no need to build up infra. Clouding computing environment become popular more and more because various computing environment using virtualization is provided. The attack threat range also becomes wider in proportion to broaden various connection ways and service supply range at these clouding computing. Therefore, intrusion detection system which can protect resource from various attack having malignant attempts is necessary. In this study, we analyzed about characteristic of intrusion detection system at cloud computing environment having big damage than other computing environment when intrusion happen by sharing of resource and virtualization.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.3
• /
• pp.952-973
• /
• 2021

In recent years, mobile devices have become an essential part of daily life. More and more applications are being supported by mobile devices thanks to edge computing, which represents an emergent architecture that provides computing, storage, and networking capabilities for mobile devices. In edge computing, heavy tasks are offloaded to edge nodes to alleviate the computations on the mobile side. However, offloading computational tasks may incur extra energy consumption and delays due to network congestion and server queues. Therefore, it is necessary to optimize offloading decisions to minimize time, energy, and payment costs. In this article, different offloading models are examined to identify the offloading parameters that need to be optimized. The paper investigates and compares several optimization techniques used to optimize offloading decisions, specifically Swarm Intelligence (SI) models, since they are best suited to the distributed aspect of edge computing. Furthermore, based on the literature review, this study concludes that a Cuckoo Search Algorithm (CSA) in an edge-based architecture is a good solution for balancing energy consumption, time, and cost.

• Communications of the Korean Mathematical Society
• /
• v.28 no.4
• /
• pp.833-850
• /
• 2013

We present an algorithm for computing the Hausdorff distance between two parametric curves in $\mathbb{R}^n$, or more generally between two sets of parametric curves in $\mathbb{R}^n$. During repeated subdivision of the parameter space, we prune subintervals that cannot contain an optimal point. Typically, our algorithm costs O(logM) operations, compared with O(M) operations for a direct, brute-force method, to achieve an accuracy of $O(M^{-1})$.