• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 춘계학술발표대회
• /
• pp.498-500
• /
• 2019

Point cloud is a flexible set of points that can provide a scalable geometric representation which can be applied in different computer graphic task. We propose a method based on EdgeConv and densely connected layers to aggregate the features for better classification. Our proposed approach shows significant performance improvement compared to the state-of-the-art deep neural network-based approaches.

• 감성과학
• /
• 제21권3호
• /
• pp.83-102
• /
• 2018

The process model of emotion regulation suggests that cognitive reappraisal and expressive suppression engage at different time points in the regulation process. Although multiple brain regions and networks have been identified for each strategy, no articles have explored changes in network characteristics or network connectivity over time. The present study examined (a) the whole-brain network and six other resting-state networks, (b) their modularity and global efficiency, which is an index of the efficiency of information exchange across the network, (c) the degree and betweenness centrality for 160 brain regions to identify the hub nodes with the most control over the entire network, and (d) the intra-network and inter-network functional connectivity (FC). Such investigations were performed using a traditional large-scale FC analysis and a relatively recent sliding window correlation analysis. The results showed that the right inferior orbitofrontal cortex was the hub region of the whole-brain network for both strategies. The present findings of temporally altering functional activity of the networks revealed that the default mode network (DMN) activated at the early stage of reappraisal, followed by the task-positive networks (cingulo-opercular network and fronto-parietal network), emotion-processing networks (the cerebellar network and DMN), and sensorimotor network (SMN) that activated at the early stage of suppression, followed by the greater recruitment of task-positive networks and their functional connection with the emotional response-related networks (SMN and occipital network). This is the first study that provides neuroimaging evidence supporting the process model of emotion regulation by revealing the temporally varying network efficiency and intra- and inter-network functional connections of reappraisal and suppression.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2003년도 ICCAS
• /
• pp.1051-1055
• /
• 2003

Policies are collections of general principles specifying the desired behavior and state of a system. Network management is mainly carried out by following policies about the behavior of the resources in the network. Policy-based (PB) network management supports to manage distributed system in a flexible and dynamic way. This paper focuses on configuration management based on Internet Engineering Task Force (IETF) standards. Network security approaches include the usage of intrusion detection system to detect the intrusion, building firewall to protect the internal systems and network. This paper presents how the policy-based framework is collaborated among the network security systems (intrusion detection system, firewall) and intrusion detection systems are cooperated to detect the intrusions.

• 정보처리학회논문지A
• /
• 제11A권3호
• /
• pp.181-188
• /
• 2004

인터넷 기반 분산/병렬 처리 시스템인 PDP(Parallel/Distributed Processing)는 인터넷의 유휴상태 호스트들을 이용하여 대용량 작업을 병렬로 처리해서 전체 수행 시간을 감소시킨다. 본 연구에서는 실시간 네트워크 모니터링을 활용하여 수시로 변화하는 네트워크 환경에 적응하여 병렬/분산 처리되는 방안을 제안한다. 실시간 네트워크 모니터링 정보를 PDP 주요 핵심 알고리즘들에 적용하여 네트워크 과부하 및 결함으로 발생하는 작업 지연 요소에 적응적으로 대처함으로써 전체 성능이 향상됨을 보인다.

• 한국멀티미디어학회논문지
• /
• 제3권3호
• /
• pp.320-332
• /
• 2000

웹은 네트워크로 연결된 모든 컴퓨터를 하나로 묶는 거대한 가상 시스템을 구성한다. 인터넷에 존재하는 수많은 유휴 상태 시스템을 이용하여 병렬 처리함으로써 비용 대 성능비가 매우 높으며 강력한 컴퓨팅 파워를 요구하는 거대한 문제를 해결할 수 있다. 하지만, 로컬 네트워크가 아닌 인터 넷 전체를 대상으로 하는 글로벌 환경에서 병렬 수행하는데 호스트들간의 이질성, 접근의 용이성, 작업에 대한 신뢰성을 고려해야 한다. 본 논문은 가상 병렬 처리 시스템인 WebImg를 설계 및 구현하여 웹 컴퓨팅 이 가능하며 동일한 작업을 여러 호스트에게 분배하기 위한 유연성 있는 작업 할당 전략을 제시하고 그 성능을 평가한다. 작업에 참여한 이 기종 호스트들이 가변적인 환경에서 작업 수행 도중 시스템의 성능변화에 대처하여 재할당 연산을 이용한 유연성 있는 작업 할당 기법을 제시한다. 더욱이 제안한 작업 할당 전략은 참여 호스트의 상태를 수시로 제어하여 결함내성을 제공한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권11호
• /
• pp.5269-5286
• /
• 2018

Task scheduling is one of the key issues in improving system performance and optimizing resource management in cloud computing environment. In order to provide appropriate services for heterogeneous users, we propose a novel task scheduling strategy with service differentiation, in which the delay sensitive tasks are assigned to the rapid cloud with high-speed processing, whereas the fault sensitive tasks are assigned to the reliable cloud with service restoration. Considering that a user can receive service from either local SaaS (Software as a Service) servers or public IaaS (Infrastructure as a Service) cloud, we establish a hybrid queueing network based system model. With the assumption of Poisson arriving process, we analyze the system model in steady state. Moreover, we derive the performance measures in terms of average response time of the delay sensitive tasks and utilization of VMs (Virtual Machines) in reliable cloud. We provide experimental results to validate the proposed strategy and the system model. Furthermore, we investigate the Nash equilibrium behavior and the social optimization behavior of the delay sensitive tasks. Finally, we carry out an improved intelligent searching algorithm to obtain the optimal arrival rate of total tasks and present a pricing policy for the delay sensitive tasks.

• 센서학회지
• /
• 제30권5호
• /
• pp.290-294
• /
• 2021

Experts have designed popular and successful model architectures, which, however, were not the optimal option for different scenarios. Despite the remarkable performances achieved by deep neural networks, manually designed networks for classification tasks are the backbone of object detection. One major challenge is the ImageNet pre-training of the search space representation; moreover, the searched network incurs huge computational cost. Therefore, to overcome the obstacle of the pre-training process, we introduce a network adaptation technique using a pre-trained backbone model tested on ImageNet. The adaptation method can efficiently adapt the manually designed network on ImageNet to the new object-detection task. Neural architecture search (NAS) is adopted to adapt the architecture of the network. The adaptation is conducted on the MobileNetV2 network. The proposed NAS is tested using SSDLite detector. The results demonstrate increased performance compared to existing network architecture in terms of search cost, total number of adder arithmetics (Madds), and mean Average Precision(mAP). The total computational cost of the proposed NAS is much less than that of the State Of The Art (SOTA) NAS method.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.217-223
• /
• 2022

In modern business conditions, the improvement of business processes cannot do without digitalization. Digital technologies allow businesses to conquer markets, quickly introduce new technologies not only into production processes, but also at all levels of economic activity. The rapid pace of development of information, communication and economic spheres determine the relevance of the research topic and the goals that digital management solves. Today, the use of digital equipment and platforms makes it possible to form the basis for the formation of competitive business advantages, minimize costs, and most importantly, respond in time to changes in both the internal and external environment..Thus, the main task of the study is to analyze the digital management system in a business environment. As a result of the study, current trends and prerequisites for digital management system in a business environment were investigated.

• International Journal of Computer Science & Network Security
• /
• 제22권6호
• /
• pp.246-250
• /
• 2022

The article analyzes the regulations of current criminal law and current issues, combating drug crime on the Internet, as well as measures to combat drug crime in the field of modern information technology. In connection with the growth of crimes in the field of drug trafficking committed with the use of information and telecommunications technologies, the urgent task of the state is to find effective ways to reduce drug crime. The article considers criminologically significant aspects of the mechanism of illicit drug trafficking, which is carried out with the use of information and telecommunication technologies and means of remote communication.

• ETRI Journal
• /
• 제42권1호
• /
• pp.90-100
• /
• 2020

Named entity recognition (NER) continues to be an important task in natural language processing because it is featured as a subtask and/or subproblem in information extraction and machine translation. In Urdu language processing, it is a very difficult task. This paper proposes various deep recurrent neural network (DRNN) learning models with word embedding. Experimental results demonstrate that they improve upon current state-of-the-art NER approaches for Urdu. The DRRN models evaluated include forward and bidirectional extensions of the long short-term memory and back propagation through time approaches. The proposed models consider both language-dependent features, such as part-of-speech tags, and language-independent features, such as the "context windows" of words. The effectiveness of the DRNN models with word embedding for NER in Urdu is demonstrated using three datasets. The results reveal that the proposed approach significantly outperforms previous conditional random field and artificial neural network approaches. The best f-measure values achieved on the three benchmark datasets using the proposed deep learning approaches are 81.1%, 79.94%, and 63.21%, respectively.