• Journal of Communications and Networks
• /
• 제17권2호
• /
• pp.145-156
• /
• 2015

In recent years, electronic commerce and online social networks (OSNs) have experienced fast growth, and as a result, recommendation systems (RSs) have become extremely common. Accuracy and robustness are important performance indexes that characterize customized information or suggestions provided by RSs. However, nefarious users may be present, and they can distort information within the RSs by creating fake identities (Sybils). Although prior research has attempted to mitigate the negative impact of Sybils, the presence of these fake identities remains an unsolved problem. In this paper, we introduce a new weighted link analysis and influence level for RSs resistant to Sybil attacks. Our approach is validated through simulations of a broad range of attacks, and it is found to outperform other state-of-the-art recommendation methods in terms of both accuracy and robustness.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2001년도 하계학술대회 논문집 D
• /
• pp.2089-2091
• /
• 2001

In this paper, a hybrid controller that consists of a conventional PD controller and a neural network controller which adapts to various control conditions by online learning is used and a new learning algorithm of the neural networks is used to prevent weights of neural network from diverging. A conventional PI controller and the hybrid controller is applied to speed control of 3 phase induction motor. So in comparison with a PD controller, we prove superiority of hybrid controller by experiments.

• 정보통신설비학회논문지
• /
• 제9권1호
• /
• pp.24-28
• /
• 2010

The Quality of Experience (QoE) of next Generation wired/wireless network services based upon IP networking is becoming a popular issue in recent years. The user experience of Internet services such as IPTV, online game, web surfing and etc, are becoming the most desirable factors to service providers to improve service performance and customer's satisfaction. However, collecting user experience from customers and obtaining the QoE parameters from the Quality of Service (QoS) parameters such as bandwidth, delay, jitter or admission control algorithm, are difficult subjects because of the various service types and user characteristics. In this paper, we propose a framework which contains service classification, QoE analysis and service enhancement steps for a suitable QoE measurement and management protocol. We define the user satisfaction indicators of the Internet services, classify the categories of each type of services, and analyse the Key Performance Indicator (KPI) in each type of services to perform the QoS parameters and improving the service qualities.

• Journal of Power Electronics
• /
• 제16권1호
• /
• pp.1-8
• /
• 2016

Dead time is typically incorporated in voltage source inverter systems to prevent short circuit cases. However, dead time causes an error between the output voltage and reference voltage. Hence, voltage equation-based algorithms, such as motor parameter estimation and back electromotive force (EMF)-based sensorless algorithms, are prone to estimation errors. Several dead-time compensation methods have been developed to reduce output voltage errors. However, voltage errors are still common in zero current crossing areas, and an effect of the error is much worse in a low speed region. Therefore, employing voltage equation-based algorithms in low speed regions is difficult. This study analyzes the conventional dead-time compensation method and output voltage errors in low speed operation areas. A current shaping method that can reduce output voltage errors is also proposed. Experimental results prove that the proposed method reduces voltage errors and improves the accuracy of the parameter estimation method and the performance of the back EMF-based sensorless algorithm.

• 한국항해항만학회지
• /
• 제31권7호
• /
• pp.563-568
• /
• 2007

In this paper, an adaptive neural network controller and its application to automatic berthing control of ship is presented. The neural network controller is trained online using adaptive interaction technique without any teaching data and off-line training phase. Firstly, the neural networks used to control rudder and propeller during automatic berthing process are presented. Secondly, computer simulations of automatic ship berthing are carried out in Pusan bay to verify the proposed controller under the influence of wind disturbance and measurement noise. The results of simulation show good performance of the developed berthing control system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제6권12호
• /
• pp.3197-3218
• /
• 2012

Compared with traditional itinerary planning, intention-oriented itinerary recommendations can provide more flexible activity planning without requiring the user's predetermined destinations and is especially helpful for those in unfamiliar environments. The rank and classification of points of interest (POI) from location-based social networks (LBSN) are used to indicate different user intentions. The mining of vehicles' physical trajectories can provide exact civil traffic information for path planning. This paper proposes a POI category-based itinerary recommendation framework combining physical trajectories with LBSN. Specifically, a Voronoi graph-based GPS trajectory analysis method is utilized to build traffic information networks, and an ant colony algorithm for multi-object optimization is implemented to locate the most appropriate itineraries. We conduct experiments on datasets from the Foursquare and GeoLife projects. A test of users' satisfaction with the recommended items is also performed. Our results show that the satisfaction level reaches an average of 80%.

• Journal of Multimedia Information System
• /
• 제5권2호
• /
• pp.99-104
• /
• 2018

Phishing website has become a crucial concern in cyber security applications. It is performed by fraudulently deceiving users with the aim of obtaining their sensitive information such as bank account information, credit card, username, and password. The threat has led to huge losses to online retailers, e-business platform, financial institutions, and to name but a few. One way to build anti-phishing detection mechanism is to construct classification algorithm based on machine learning techniques. The objective of this paper is to compare different classifier ensemble approaches, i.e. random forest, rotation forest, gradient boosted machine, and extreme gradient boosting against single classifiers, i.e. decision tree, classification and regression tree, and credal decision tree in the case of website phishing. Area under ROC curve (AUC) is employed as a performance metric, whilst statistical tests are used as baseline indicator of significance evaluation among classifiers. The paper contributes the existing literature on making a benchmark of classifier ensembles for web phishing detection.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2014년도 춘계학술발표대회
• /
• pp.1026-1029
• /
• 2014

MOBA (Multiplayer Online Battle Arena) 기반의 게임 서비스는 현재 가장 관심을 받고 있는 게임 장르의 한 종류이다. MOBA 장르와 같은 게임들은 플레이어의 실력도 중요하지만 같은 팀원간의 협력과 전략이 중요한 요소 중에 하나이다. 이러한 상황에서 악의적의 의도로 자신의 비정상적인 플레이를 한다거나 욕설 등의 팀의 사기를 저하시키는 플레이어들이 문제가 되고 있다. 이러한 플레이어들의 제재를 위해 몇 가지 시스템들이 제안 되고 있지만 그들에 대한 판단은 쉽지 않다. 그래서 본 논문에서는 PageRank 를 기반으로 하는 불량 플레이어의 판단에 대한 보조 시스템을 제안 한다. 이 시스템의 MOBA 게임 플레이어의 플레이 횟수, 신고 횟수, 신고 받은 횟수 등의 자료들을 이용하여 플레이어의 Judgment Points 와 Bad Player 지수를 파악하며 이를 기반으로 생성된 Bad Player 랭킹을 통하여 불량 플레이어 검색에 도움을 줄 것으로 예상된다.

• 한국벤처창업학회:학술대회논문집
• /
• 한국벤처창업학회 2019년도 추계학술대회
• /
• pp.135-141
• /
• 2019

최근 대표적 글로벌 유통기업인 미국의 아마존과 중국의 알리바바가 전 세계적으로 가장 큰 시장점유가 있으며 두 기업의 국내진입 시 국내 유통산업에 큰 영향을 미칠 것으로 예상한다. 두 기업은 온라인 기업이 오프라인 기업을 흡수 합병함으로써 새로운 가치를 창출해내는 O2O (Online to Offline) 추세가 국제적으로 진행되고 있다. 아마존과 알리바바와 같은 글로벌 유통업체들은 일본, 인도와 같은 타 국가로의 세계 진출을 적극적으로 하는 추세이다. 본 연구에서는 아마존, 알리바바와 같은 글로벌 유통업체가 세계 진출의 일환으로 국내 유통시장 진입 시, 노출된 글로벌 경쟁 속에서 국내 유통기업들의 사업전망을 예측해보고, 해당 예측에 기반하여 기업 차원의 전략적 대응방안 및 정부 차원의 정책 지원방안을 마련하는 데 그 목적이 있다. 시장 현황분석을 기반으로 하여, 미래 시장예측 방법으로써 무작위로 추출된 난수(Random Number)를 이용하여 원하는 방정식의 값을 확률적으로 구하기 위한 알고리즘(Algorithm) 및 시뮬레이션(Simulation)의 방법인 몬테카를로(Monte Carlo, MC) 방법론을 사용하여 국내 유통시장의 변화를 예측하여 본 연구를 진행하였다.

• 한국컴퓨터정보학회논문지
• /
• 제23권1호
• /
• pp.57-66
• /
• 2018

It is known that a single-key and a related-key attacks on AES-128 are possible for at most 7 and 8 rounds, respectively. The security of CMAC, a typical block-cipher-based MAC algorithm, has very high possibility of inheriting the security of the underlying block cipher. Since the attacks on the underlying block cipher can be applied directly to the first block of CMAC, the current security margin is not sufficient compared to what the designers of AES claimed. In this paper, we consider HMAC-DM-AES-128 as an alternative to CMAC-AES-128 and analyze its security for reduced rounds of AES-128. For 2-round AES-128, HMAC-DM-AES-128 requires the precomputation phase time complexity of $2^{97}$ AES, the online phase time complexity of $2^{98.68}$ AES and the data complexity of $2^{98}$ blocks. Our work is meaningful in the point that it is the first security analysis of MAC based on hash modes of AES.