• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.401-404
• /
• 2020

현재 전자상거래에서 사용되는 협업 필터링은 고객이 입력한 평점 정보를 이용하여 추천 시스템을 구축한다. 하지만 기존의 평점 정보는 고객이 직접 입력해야 하므로 데이터 희소생의 문제가 있고 허위정보를 가려내지 못한다는 문제점 또한 존재한다. 본 논문에서는 기존 평점 정보 기반의 협업 필터링 추천 시스템의 문제점을 해결하기 위해, 온라인 고객 행동 정보를 활용한 협업 필터링 알고리즘을 제안하였다. 실험 결과 본 연구에서 제안한 Collaborative Filtering based on Online Behavior Information (BICF) 알고리즘이 기존의 평점 기반 협업 필터링 방식보다 우수한 성능을 보임을 보여주었다.

• Journal of KIISE:Software and Applications
• /
• v.33 no.5
• /
• pp.508-524
• /
• 2006

We present our experience of combining, in a realistic setting, a static analyzer with a statistical analysis. This combination is in order to reduce the inevitable false alarms from a domain-unaware static analyzer. Our analyzer named Airac(Array Index Range Analyzer for C) collects all the true buffer-overrun points in ANSI C programs. The soundness is maintained, and the analysis' cost-accuracy improvement is achieved by techniques that static analysis community has long accumulated. For still inevitable false alarms (e.g. Airac raised 970 buffer-overrun alarms in commercial C programs of 5.3 million lines and 737 among the 970 alarms were false), which are always apt for particular C programs, we use a statistical post analysis. The statistical analysis, given the analysis results (alarms), sifts out probable false alarms and prioritizes true alarms. It estimates the probability of each alarm being true. The probabilities are used in two ways: 1) only the alarms that have true-alarm probabilities higher than a threshold are reported to the user; 2) the alarms are sorted by the probability before reporting, so that the user can check highly probable errors first. In our experiments with Linux kernel sources, if we set the risk of missing true error is about 3 times greater than false alarming, 74.83% of false alarms could be filtered; only 15.17% of false alarms were mixed up until the user observes 50% of the true alarms.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2020.01a
• /
• pp.37-39
• /
• 2020

인터넷 상에서의 허위정보 생산과 유통은 주로 가짜 뉴스를 통하여 이루어진다. 과거에는 신문이나 공중파 TV등 뉴스 기사의 생산과 유통이 매우 제한적이었지만 지금은 인터넷의 발달로 누구나 쉽게 뉴스를 생산하고 유통할 수 있다. 뉴스 생산의 용이성은 정보 공유의 즉각성과 수월성이라는 장점을 제공하지만 반대로 불확실한 뉴스 남발로 인한 정보의 신뢰성 하락과 선량한 피해자를 양산하는 단점 또한 존재한다. 이는 가짜 뉴스가 사회적 문제로 대두되고 있는 이유이다. 에이전트나 스파이더 등의 소프트웨어를 통해 인터넷으로 급속도로 전파되는 가짜 뉴스를 전통 방식인 소수의 전문가가 수동으로 잡아내는 것은 불가능하다. 이에 기술발달로 잡아내기 힘들어진 가짜뉴스에 대해, 역으로 발달된 기술을 활용하여 잡아내려는 시도가 늘어나고 있다. 본 논문에서는 가짜뉴스를 판별하는 다양한 기법들을 탐색하고 해결방안을 제시하고자 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.2
• /
• pp.113-119
• /
• 2016

Wireless sensor networks are vulnerable to an adversary due to scarce resources and wireless communication. An adversary can compromise a sensor node and launch a variety of attacks such as false report injection attacks. This attack may cause monetary damage resulting in energy drain by forwarding the false reports and false alarms at the base station. In order to address this problem, a number of en-route filtering schemes has been proposed. Notably, a dynamic en-route filtering scheme can save energy by filtering of the false report. In the key dissemination phase of the existing scheme, the nodes closer to the source node may not have matching keys to detect the false report. Therefore, continuous attacks may result in unnecessary energy wastage. In this paper, we propose a key re-distribution scheme to solve this issue. The proposed scheme early detects the false report injection attacks using initially assigned secret keys in the phase of the key pre-distribution. The experimental results demonstrate the validity of our scheme with energy efficiency of up to 26.63% and filtering capacity up to 15.92% as compared to the existing scheme.

• Proceedings of the Korea Contents Association Conference
• /
• 2014.11a
• /
• pp.67-68
• /
• 2014

스마트기기의 대중화와 SNS의 활용은 우리 사회생활에 많은 변화를 가져왔다. 하지만 이를 활용함으로써 나타는 폐해는 개인의 프라이버시 침해, 사이버폭력, 허위정보의 양산, 각종 범죄 등에 노출되는 부작용을 보이고 있다. 이 연구에서는 국내의 SNS 활용실태와 현재 경찰의 4대악 척결의 하나인 학교폭력을 중심으로 고찰을 시도하고자 한다. 특히 SNS의 급격한 확산으로 인해 발생되는 사이버불링(따돌림)의 실태에 대해 파악하고 그에 대한 제언을 하고자 한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.19 no.4
• /
• pp.41-49
• /
• 2009

In ubiquitous sensor networks, sensor nodes can be compromised by an adversary since they are deployed in hostile environments. False sensing reports can be injected into the network through these compromised nodes, which may cause not only false alarms but also the depletion of limited energy resource in the network. In the security solutions for the filtering of false reports, the choice of a security threshold value which determines the security level is important. In the existing adaptive solutions, a newly determined threshold value is broadcasted to the whole nodes, so that extra energy resource may be consumed unnecessarily. In this paper, we propose an application of the recursive contract net protocol to determine the threshold value which can provide both energy efficiency and sufficient security level. To manage the network more efficiently, the network is hierarchically grouped, and the contract net protocol is applied to each group. Through the protocol, the threshold value determined by the base station using a fuzzy logic is applied only where the security attack occurs on.

• Agrochemical news magazine
• /
• v.20 no.6 s.153
• /
• pp.58-61
• /
• 1999

• Journal of the Korea Society for Simulation
• /
• v.16 no.2
• /
• pp.27-35
• /
• 2007

When sensor networks are deployed in open environments, all the sensor nodes are vulnerable to physical threat. An attacker can physically capture a sensor node and obtain the security information including the keys used for data authentication. An attacker can easily inject false reports into the sensor network through the compromised node. False report can lead to not only false alarms but also the depletion of limited energy resource in battery powered sensor networks. To overcome this threat, Fan Ye et al. proposed that statistical on-route filtering scheme(SEF) can do verify the false report during the forwarding process. In this scheme, the choice of a security threshold value is important since it trades off detection power and energy, where security threshold value is the number of message authentication code for verification of false report. In this paper, we propose a fuzzy rule-based system for security threshold determination that can conserve energy, while it provides sufficient detection power in the SEF based sensor networks. The fuzzy logic determines a security threshold by considering the probability of a node having non-compromised keys, the number of compromised partitions, and the remaining energy of nodes. The fuzzy based threshold value can conserve energy, while it provides sufficient detection power.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.05a
• /
• pp.328-328
• /
• 2022

• Journal of the Korean Society for Library and Information Science
• /
• v.53 no.3
• /
• pp.291-315
• /
• 2019

The aim of the present study was to compare the research trends on disinformation between Korean and abroad. To achieve this objective, a total of 283 author-assigned English keywords in 104 Korean papers and 3,551 author-assigned English keywords in 861 abroad papers were collected from the whole research fields and the publication periods. The collected data were analyzed using NetMiner V.4 to discover their 'degree centrality' and 'betweenness centrality'of the keyword network. The result are as follows. First, the major research topics of disinformation in Korea were drawn such as 'Freedom of Expression', 'Fact Check', 'Regulation', 'Media Literacy', and 'Information Literacy' in order; whereas, in abroad were shown like 'Social Media', 'Post Truth', 'Propaganda', 'Information Literacy', and 'Journalism' in order. Second, in terms of the influence of research topics related to disinformation, in Korea were identified such as 'Fact Check', 'Freedom of Expression', and 'Hoax' in order; whereas, in abroad were shown such as 'Social Media' and 'Detection' in order. Finally, in an aspect of intervention of research topics related to disinformation, in Korea were 'Fact Check', 'Polarization', 'Freedom of Expression', and 'Commercial'; whereas, in abroad were 'Social Media', 'Detection', and 'Machine Learning' in order.