• 정보처리학회논문지D
• /
• 제16D권3호
• /
• pp.317-326
• /
• 2009

최근 궤적 정보를 이용한 많은 연구들이 진행되고 있으나, 이들 대부분의 연구는 유클리드 공간 내의 궤적들을 대상으로 하고 있다. 그러나 실제 응용에서 대부분의 이동 객체들은 도로 네트워크 공간상에 존재하므로, 유클리드 공간을 대상으로 한 연구들을 도로 네트워크 공간에 적용시키는 것은 적합하지 않다. 본 논문에서는 도로 네트워크 내 이동 객체들의 대용량 궤적 정보를 대상으로 하는 효과적인 클러스터링 기법에 대하여 논한다. 이를 위하여 우선 본 논문에서는 궤적을 각 이동 객체가 시간에 따라 지나온 도로 세그먼트들의 연속으로 정의한다. 다음, 도로 세그먼트들의 길이와 식별자 정보를 이용한 새로운 유사도 측정 함수를 제안하고, 이를 이용하여 측정된 궤적간의 유사도 정보를 기반으로 FastMap과 계층 클러스터링(hierarchical clustering)기법을 이용하여 전체 궤적들을 클러스터링하는 방식을 제안한다. 또한, 본 논문에서는 실제 응용에서 대부분의 이동 객체는 최단 거리를 이용하여 움직인다는 특성을 반영한 새로운 궤적 생성 기법을 제안하고, 이렇게 생성된 궤적 데이터를 이용하여 제안된 클러스터링 기법에 대한 다양한 성능 평가 결과를 보인다. 실험 결과에 따르면 제안된 기법은 사람에 의하여 유사 궤적들을 클러스터링한 결과와 비교하여 95%이상의 높은 정확도를 보였다.

• 한국정보과학회논문지:시스템및이론
• /
• 제26권12호
• /
• pp.1493-1503
• /
• 1999

ATM Clustering System과 같이 SAN(System Area Network) 환경에서 동작하는 시스템은 낮은 지연시간과 넓은 대역폭의 네트워크가 필수적이나 기존의 에러 복구 프로토콜들은 이러한 요구를 충족시키기에는 큰 오버헤드를 가지고 있다. 제안된 새로운 에러 복구 프로토콜은 ATM Clustering System 환경에서 최적의 성능을 나타내는 light-weight 프로토콜로 에러가 없는 상황과 에러 복구가 진행중인 상황에 따라 acknowledgement 주기를 적응적으로 변화시키는 adaptive acknowledgement scheme를 제안하여 적용하였다. 제안된 프로토콜은 상용 툴인 SDT를 이용한 논리 검증 받았고, DEVSim++ 환경에서의 성능 분석을 통해 프로토콜이 최상의 성능을 보이기 위한 파라메터 값을 찾았고, 이 값을 적용하였을 때의 성능을 기존의 프로토콜과 비교하여 제안된 프로토콜이 더 우수함을 확인하였다.Abstract While a system working with SAN, such as ATM Clustering System, requires a network with low latency and wide bandwidth, the previous error recovery protocols have a serious network overhead to satisfy this requirement. The suggested error recovery protocol is a light-weight protocol which can shows its best performance at ATM Clustering System and uses a newly suggested adaptive acknowledgement scheme. In the adaptive acknowledgement scheme, the period of acknowledgement is dynamically changed depending on the state of the network. We proved the logical correctness of our protocol with SDT and did performance analysis with DEVSim++. From the analysis, we found the optimal parameter values for best performance and showed that our protocol works better than the previous error recovery protocols.

• Journal of Computing Science and Engineering
• /
• 제10권4호
• /
• pp.103-110
• /
• 2016

Emotion recognition is an important component of affective computing, and is significant in the implementation of natural and friendly human-computer interaction. An effective approach to recognizing emotion from text is based on a machine learning technique, which deals with emotion recognition as a classification problem. However, in emotion recognition, the texts involved are usually very short, leaving a very large, sparse feature space, which decreases the performance of emotion classification. This paper proposes to resolve the problem of feature sparseness, and largely improve the emotion recognition performance from short texts by doing the following: representing short texts with word cluster features, offering a novel word clustering algorithm, and using a new feature weighting scheme. Emotion classification experiments were performed with different features and weighting schemes on a publicly available dataset. The experimental results suggest that the word cluster features and the proposed weighting scheme can partly resolve problems with feature sparseness and emotion recognition performance.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2006년 학술대회 논문집 정보 및 제어부문
• /
• pp.366-368
• /
• 2006

Electronic warfare Support(ES) system measures pulse characteristics for received RF signals that received from all directions. ES system discriminates the pulse trains that have a rule, correlationship, continuance from collected data and analyze the characteristics of the data, and identify the emitters by comparison with emitter identification data(EID). Because pulse density is very high and various signal source exists at modem signal environments, high-speed and accurate signal analysis is needed for realtime countermeasure to emitters. Grouping alleviates the load of signal analysis process and supports reliable analysis. In this paper, we suggest an adaptive clustering scheme regarding signal patterns.

• 대한자기공명의과학회:학술대회논문집
• /
• 대한자기공명의과학회 2003년도 제8차 학술대회 초록집
• /
• pp.34-34
• /
• 2003

Age-related WMCs frequently appear in older subjects and are known to be associated with cognitive impairment and brain pathologies such as Alzheimer's disease and stroke. However, it is difficult to detect WMC correctly by using only intensity-based clustering scheme because the intensity levels of WC are similar to those of gray matter(GM). In this paper, we aimed to develop a fast and accurate scheme to detect and segment periventricular WMCs by using both k-means clustering method and morphological features.

• Journal of Communications and Networks
• /
• 제17권5호
• /
• pp.473-481
• /
• 2015

In this study, we propose a clustering algorithm to enhance the performance of wireless sensor and actuator networks (WSANs). In each cluster, a multi-level hierarchical structure can be applied to reduce energy consumption. In addition to the cluster head, some nodes can be selected as intermediate nodes (INs). Each IN manages a subcluster that includes its neighbors. INs aggregate data from members in its subcluster, then send them to the cluster head. The selection of intermediate nodes aiming to optimize energy consumption can be considered high computational complexity mixed-integer linear programming. Therefore, a heuristic lowest energy path searching algorithm is proposed to reduce computational time. Moreover, a channel assignment scheme for subclusters is proposed to minimize interference between neighboring subclusters, thereby increasing aggregated throughput. Simulation results confirm that the proposed scheme can prolong network lifetime in WSANs.

• Journal of Electrical Engineering and information Science
• /
• 제2권6호
• /
• pp.191-196
• /
• 1997

Fuzzy clustering has been playing an important role in solving many problems. Fuzzy c-Means(FCM) algorithm is most frequently used for fuzzy clustering. But some fixed point of FCM algorithm, know as Tucker's counter example, is not a reasonable solution. Moreover, FCM algorithm is impossible to perform the on-line learning since it is basically a batch learning scheme. This paper presents unsupervised learning networks as an attempt to improve shortcomings of the conventional clustering algorithm. This model integrates optimization function of FCM algorithm into unsupervised learning networks. The learning rule of the proposed scheme is a result of formal derivation based on the gradient descent procedure of a fuzzy objective function. Using the result of formal derivation, two algorithms of fuzzy cluster analysis, the batch learning version and on-line learning version, are devised. They are tested on several data sets and compared with FCM. The experimental results show that the proposed algorithms find out the reasonable solution on Tucker's counter example.

• Journal of Information Processing Systems
• /
• 제7권1호
• /
• pp.29-42
• /
• 2011

Wireless sensor networks are composed of a large number of sensor nodes with limited energy resources. One critical issue in wireless sensor networks is how to gather sensed information in an energy efficient way, since their energy is limited. The clustering algorithm is a technique used to reduce energy consumption. It can improve the scalability and lifetime of wireless sensor networks. In this paper, we introduce a clustering protocol with mode selection (CPMS) for wireless sensor networks. Our scheme improves the performance of BCDCP (Base Station Controlled Dynamic Clustering Protocol) and BIDRP (Base Station Initiated Dynamic Routing Protocol) routing protocol. In CPMS, the base station constructs clusters and makes the head node with the highest residual energy send data to the base station. Furthermore, we can save the energy of head nodes by using the modes selection method. The simulation results show that CPMS achieves longer lifetime and more data message transmissions than current important clustering protocols in wireless sensor networks.

• Journal of information and communication convergence engineering
• /
• 제5권2호
• /
• pp.104-108
• /
• 2007

Ad hoc networks are a type of mobile network that function without any fixed infrastructure. One of the weaknesses of ad hoc network is that a route used between a source and a destination is to break during communication. To solve this problem, one approach consists of selecting routes whose nodes have the most stable link cost. This paper proposes a method for improving the low power distributed MAC. This method is based on the context awareness of the each nodes energy in clustering. We propose to select a new scheme to optimize energy conserving between the clustering nodes in MANET. And this architecture scheme would use context-aware considering the energy related information such as energy, RF strength, relative distances between each node in mobile ad hoc networks. The proposed networks scheme could get better improve the awareness for data to achieve and performance on their clustering establishment and messages transmission. Also, by using the context aware computing, according to the condition and the rules defined, the sensor nodes could adjust their behaviors correspondingly to improve the network routing.

• 한국ITS학회 논문지
• /
• 제7권6호
• /
• pp.101-109
• /
• 2008

본 논문은 클러스터링 기법을 도입하여 기존에 제안된 가중치 협력 스펙트럼 시스템에서 실질적으로 구하지 못했던 Pd를 구하고, 새로운 가중치 생성 알고리즘을 통하여 1차 사용자 신호의 감지 성능을 향상시키는 방법을 제안하였다. 유사한 채널을 같는 CR 사용자를 클러스터링 기법을 이용하여 그룹화하여 각각의 사용자로부터 획득한 센싱 결과를 토대로 Pd를 계산하였다. 또한, 각 클러스터의 검출확률의 제곱 합을 이용하여 가중치(Wj(n+1))를 생성하였다. 이는 기존의 방식보다 센싱 성능이 우수하였으며, 특히 1차 사용자의 신호가 갑자기 사라졌을 경우 신호가 없는 상황에서의 검출 확률인 false alarm rate가 낮아지는 결과를 보였다. 컴퓨터 모의실험을 통하여 이를 검증한다.