• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.11
• /
• pp.1116-1120
• /
• 2010

In this paper we present two refined neighbor selection algorithms for recommender systems and also show how the attributes of the items can be used for higher prediction quality. The refined neighbor selection algorithms adopt the transitivity-based neighbor selection method using virtual neighbors and alternate neighbors, respectively. The experimental results show that the recommender systems with the proposed algorithms outperform other systems and they can overcome the large scale dataset problem as well as the first rater problem without deteriorating prediction quality.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.5
• /
• pp.809-818
• /
• 2009

Collaborative filtering approach predicts the preference of active user about specific items transacted on the e-commerce by using others' preference information. To improve the prediction accuracy through collaborative filtering approach, it must be needed to gain enough preference information of users' for predicting preference. But, a bit much information of users' preference might wrongly affect on prediction accuracy, and also too small information of users' preference might make bad effect on the prediction accuracy. This research suggests the method, which decides suitable numbers of neighbor users for applying collaborative filtering algorithm, improved by existing k nearest neighbors selection methods. The result of this research provides useful methods for improving the prediction accuracy and also refines exploratory data analysis approach for deciding appropriate numbers of nearest neighbors.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.166-169
• /
• 2014

무선 센서 네트워크는 제한된 에너지를 가진 센서 노드들로 구성되며, 센서 노드의 에너지를 효율적으로 활용하기 위해 클러스터링 알고리즘을 사용한다. 균형 있는 클러스터 구성을 위해서는 클러스터 헤드의 선정이 중요하다. 기존의 연구는 확률, 노드의 잔여 에너지, 이웃 노드의 수, 이웃 노드와의 거리 등의 정보를 활용하여 클러스터 헤드를 선정하였다. 그러나 확률은 클러스터 헤드의 밀집으로 인한 에너지 소비의 불균형이 있을 수 있으며, 이웃 노드와의 정보 비교는 필요한 정보 수집을 위해 많은 에너지가 필요하다. 이러한 문제점을 개선하기 위해 본 논문은 센서 노드를 베이스 스테이션과의 거리에 따라 2-레벨로 나누고 각 상위 레벨에 속한 동일한 하위 레벨을 순차적으로 변경해가며 클러스터를 구성하는 기법을 제안한다.

• The KIPS Transactions:PartD
• /
• v.14D no.3 s.113
• /
• pp.347-354
• /
• 2007

It is not easy for the customers to search the valuable information on the goods among countless items available in the Internet. In order to save time and efforts in searching the goods the customers want, it is very important for a recommender system to have a capability to predict accurately customers' preferences. In this paper we present a refined neighbor selection algorithm for clustering based collaborative filtering in recommender systems. The algorithm exploits a graph approach and searches more efficiently for set of influential customers with respect to a given customer; it searches with concepts of weighted similarity and ranked clustering. The experimental results show that the recommender systems using the proposed method find the proper neighbors and give a good prediction quality.

• Journal of the Korea society of information convergence
• /
• v.7 no.2
• /
• pp.59-65
• /
• 2014

Grassfire spot matching method is based on similarity comparison of topological patterns for neighbor spots. This is a method where spot matching is performed as if fire spreads all around on grass. Spot matching starts from a seed spot pair confirmed as a matched pair of spots and spot matching spreads to the direction where the best matching result is produced. In this paper, it is a bit complicated way of grassfire method where multi-seed matched spot pair are manually selected and spot matching is performed from each multi-seed matched spot pair. The proposed method shows better performance in detection rate and accuracy than that of the previous method.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.3
• /
• pp.608-614
• /
• 2009

In wireless sensor networks, there are hundreds to thousands of small battery powered devices which are called sensors. As sensors have a limited energy resources, there is a need to use it effectively. A clustering based routing protocol forms clusters by distributed algorithm. Member nodes send their data to their cluster heads then cluster heads integrate data and send to sink node. In this paper we propose an energy efficient cluster-head selection algorithm. We have used some factors(a previous cluster head experience, a existence of data to transmit and an information that neighbors have data or not) to select optimum cluster-head and eventually improve network lifetime. Our simulation results show its effectiveness in balancing energy consumption and prolonging the network lifetime compared with LEACH and HEED algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.38A no.12
• /
• pp.1054-1060
• /
• 2013

This paper proposes a vehicle ad-hoc network (VANET) routing algorithm to reduce the probability of routing path discontinuity in urban intersections. In urban intersections, the vehicles may frequently change their moving directions. It can cause the disconnection of routing path and the increase of a packet transmission delay. In order to resolve this problem, the proposed routing algorithm exploits the information of moving directions in urban intersections. In this way, the proposed algorithm can reduce the probability of the local maximum which causes the increase of the number of routing hops and packet transmission delay. Simulation results show that the proposed algorithm can reduce the local maximum probability by 10% and increase the successful packet transmission ratio by 5% compared to the conventional VANET routing algorithms.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.4A
• /
• pp.351-357
• /
• 2007

Numerous papers that study ad hoc networks have used a hierarchical network structure to enhance scalability. The hierarchical structure typically consists of a number of clusters, each of which has its own clusterhead that maintains information. Clusterheads often need to exchange information among themselves in order to maintain information, and for such cases, a mechanism is needed to efficiently deliver information from one clusterhead to another. Here, we proposed a new distributed algorithm in which every node independently makes the decision about whether or not it forwards a received message. We used a simulation to demonstrate that the algorithm developed for this study is a considerable improvement over the control overhead algorithm.

• Proceedings of the Safety Management and Science Conference
• /
• 2012.11a
• /
• pp.87-92
• /
• 2012

본 연구는 전자상거래에서 협력적 필터링 알고리즘을 통한 사용자의 선호도 예측 정확도와 사용자가 평가한 선호도 평가치의 관계를 분석하여 알고리즘의 예측 정확도에 영향을 미치는 평가치의 통계적 특성에 관하여 연구한다. 협력적 필터링 알고리즘의 예측 정확도는 상품에 대해 공통의 관심을 갖는 이웃 사용자들의 선정과 이들의 선호도 경향이 중요한 요인이지만 본 연구에서는 선호도 예측을 위한 자신의 선호도 평가치 특성이 알고리즘에 중요한 요인임을 제시한다. 이러한 평가치의 평균, 표준편차, 왜도, 첨도 등과 같은 통계적 특성이 선호도 예측 정확도와 연관성이 있음을 제시하여 차후 연구에서 선호도 예측 이전에 사용자의 선호도 예측성과에 대한 사전평가의 가능성을 제시하고자 한다.

• Proceedings of the Korea Inteligent Information System Society Conference
• /
• 2007.11a
• /
• pp.324-328
• /
• 2007

본 논문은 감시정찰 센서네트워크에서 센서노드의 주요 기능인 표적의 탐지 및 식별을 위한 알고리즘을 제안한다. 감시정찰 센서네트워크에서 각 센서노드는 노드의 크기 및 센서, 프로세서, 네트워크, 전원 등의 자원의 제약이 있기 때문에 침입하는 적의 탐지 및 종류 식별을 위해서는 효율적인 알고리즘의 선정과 최적화가 요구된다. 본 논문에서는 음향, 진동, PIR, 자기 센서 등을 이용하여 사람, 차량 및 궤도 차량의 침입을 탐지하기 위한 적응 임계값 알고리즘과 그 종류를 식별하기 위한 최대우도추정 기법, k-최근접 이웃 추정 기법에 기반한 표적의 탐지 및 식별 알고리즘을 제안한다. 실험결과 음향 및 진동 센서에 의한 차량의 탐지, PIR 센서에 의한 사람의 탐지가 가능함을 확인할 수 있었으며 주파수 특징점을 이용하여 차량과 궤도차량의 종류식별이 가능함을 확인할 수 있었다.