A5101835240- Recommender Systems and Techniques
- Education and Learning Interventions
- Distributed systems and fault tolerance
- Educational Systems and Policies
- Petri Nets in System Modeling
- Interconnection Networks and Systems
- Mobile Crowdsensing and Crowdsourcing
- Psychosocial Factors Impacting Youth
- Educational Research and Pedagogy
- Education, Safety, and Science Studies
- Advanced Bandit Algorithms Research
- Image Retrieval and Classification Techniques
- Parallel Computing and Optimization Techniques
- Technology Adoption and User Behaviour
- Speech and dialogue systems
- Image and Video Quality Assessment
- Technology and Data Analysis
- Digital Marketing and Social Media
- Evacuation and Crowd Dynamics
- Expert finding and Q&A systems
- Corporate Social Responsibility Reporting
- Human Mobility and Location-Based Analysis
- Authorship Attribution and Profiling
- Text and Document Classification Technologies
- Job Satisfaction and Organizational Behavior
Gyeongin National University of Education
2006-2024
Samsung (South Korea)
1995-2024
University of Iowa
2005
Incheon National University
2001-2003
University of Education, Winneba
2002
Texas A&M University
1995-2002
We investigated how employees' perceptions of the internal (endorsed by firm and in line with its mission values) external (driven practical or instrumental benefits) legitimacy their firm's corporate social responsibility (CSR) activities influence work orientations. Specifically, we believed that would be considered more authentic than legitimacy, it have a positive on organizational identification affective attachment. conducted field survey using employee data from 38 firms listed Dow...
In the literature, only a few studies have been performed on distributed deadlock detection and resolution problem in generalized request model. Most of are based diffusing computation technique where propagation probes backward replies required to detect deadlock. The carry dependency information between processes for initiator algorithm determine Since fast is critical, we take centralized approach that removes need replies, but sends directly algorithm. This enables reduction time cost...
The paper presents a probabilistic performance analysis of deadlock detection algorithm in distributed systems. Although there has been extensive study on algorithms systems, little attention paid to the these algorithms. Most work achieved through simulation but not an analytic model. Min (1990), best our knowledge, made sole attempt evaluate analytically. Being different from Min's, approach takes time-dependent behavior each process into consideration rather than simply taking mean-value...
Presents a distributed algorithm for detecting generalized deadlocks in systems. The constructs spanning tree through the propagation of probes and receiving replies from those probes. initiator builds local wait-for graph to determine existence deadlock. A scheme encoding path information each process is developed so that ancestor-descendant relationships between nodes are not explicitly sent but inferred at initiator. advantages proposed are: (1) all reachable resolved, whereas current...
We propose a distributed deadlock detection algorithm which detects cycles made of back edges in search tree. Our the deadlocks initiator is directly or indirectly involved, while most proposed algorithms only detect involved. reduced version one previously, resolves all reachable from single execution by propagating probe messages to nodes tree and having each node resolve its subtree with collected information on dependency between nodes. substantially reduces message overhead previous...
This paper presents an algorithm for detecting generalized deadlocks in distributed systems. Most of the existing algorithms use a diffusing computation technique deadlock detection which messages called probes are propagated through wait-for graph and replies sent backwards to initiator carrying information on relations between nodes. Based collected information, determines deadlock. In order expedite activity, proposed exploits rather than carry required detection. As propagated,...
Objective: The present study was to improve the graphic user interface (GUI) of a military computer-based training (CBT) system in terms usability.BR Background: Existing studies have focused on usability evaluation particular GUI type such as sequence, hierarchy, or context type; however, few research has been conducted which identifies preferred features based comparative analysis different types.BR Method: A by 9 CBT design experts using 7-point scale (1: very low, 4: neutral, and 7:...
Deadlock detection and recovery-based routing schemes for wormhole networks have gained attraction because unlike deadlock avoidance-based schemes, they do not restrict adaptability. In order to alleviate the overhead of running a recovery procedure, studies on focused accuracy detection, trying reduce number false detections. This paper proposes both schemes. The proposed scheme is based turn model designed declare only one packet per simple cycle blocked packets as deadlocked. Our adjusts...
The increasing demand for higher memory capacity has led to a greater emphasis on high-density DIMM such as 3-dimension stacked (3DS) 128GB products compared the current mainstream 64GB DIMM. However, adoption of 3DS-DDR5 necessitates use technologies like through-silicon via (TSV), which can lead potential drawbacks price escalation and additional read latency. To overcome these challenges, there exists need explore development high-capacity using monolithic die approach. Furthermore, high...
연구 목적: 본 연구에서는 온라인 아동·청소년 성착취 범죄양상의 최근 동향과 플랫폼 규제에 대한 국제적인 흐름을 파악하고자 한다.연구 방법: 정보통신기술의 발전과 함께 진화된 아동 성학대 및 범죄양상을 1인칭 성학대물, 인터넷을 이용한 성적 목적의 인신매매, 실시간 성학대로 나누어 고찰하고, 이에 효과적으로 대응하기 위한 국내외 규제의 방향성과 한계를 살펴본다.연구 내용: 성착취물 문제를 해결하기 위해선 규제가 필수적이며, 특히 사업자의 유통방지에 책임 의무가 강조되고 있다. 그러나, 서비스에 단대단 암호화 기술이 탑재되는 경우가 증가하면서, 사업자와 수사기관 모두 대화 당사자 간의 내용에 접근할 수 없게 되었다. 이 상황을 악용하는 범죄자가 증가함에 따라, 국제적으로는 새로운 방향성이 논의되고 특히, 사생활 침해 등 개인의 기본권 침해를 최소화하는 동시에 수사기관이 원활한 수사를 위해 암호화된 정보에 액세스할 있도록 하는 균형점을 모색해야 한다. 다만, 이를 현재로서는...
A time-aware collaborative filtering-based recommender system provides item recommendations for the current user by prioritizing recent items preferred similar neighbors over past items. The similarity measure critically affects performance of system, and this study focuses on measuring between users that changes time. After dividing users’ rating time into intervals computing each interval, final is generated as a weighted sum assigning lower weights to values higher more values....
Abstract Current web personalization mainly focuses on extracting pages interesting to users. However, difficulty in understanding contents often arises for some users due the lack of background knowledge contents. As a solution this problem, paper proposes method personalizing search that produces results suitable level user. Fuzzy sets and membership functions are defined order keep adjusting user, which is updated based levels user shows preferences. Extensive simulation experiments...
Collaborative filtering-based recommender systems have been widely used in many commercial systems, giving a great aid selecting products to users. These however, still much be improved terms of prediction or recommendation accuracy. This paper intends improve the previous similarity measures find neighbor users which combine heuristic weights with traditional measures. The proposed method exploits information entropy and incorporates it into measures, so that global rating behavior on items...
Ramp metering is an important technique to effectively utilize the freeway capacity by regulating traffic flow of entrance ramps. In this extended abstract, we discuss outline free control algorithm Advanced distributed Metering System, ARMS, for real-time ramp systems. ARMS consists three levels algorithms integrated together control, breakdown prediction, and congestion resolution. Based on a model, takes into account risk when it tries maximize throughput. The prediction can make...
Deadlock detection and recovery-based routing protocols in wormhole networks have gained attraction because they do not restrict adaptability unlike deadlock avoidance-based protocols. Network performance largely relies on the accuracy of detection. The lower number packets presumed as deadlocked by a protocol, better network performs, since rarely enters into state reality those are usually killed or recovered according to recovery procedure, causing extra overhead network. This paper...