• 한국통신학회논문지
• /
• 제29권5C호
• /
• pp.700-705
• /
• 2004

Autoencoder와 Fuzzy c-Means 알고리즘을 이용하여, 불완전한 데이터의 군집화를 위한 알고리즘이 본 논문에서 제안되었다. 본 논문에서 제안된 Optimal Completion Autoencoder Fuzzy c-Means (OCAEFCM)은 손상되어 불완전한 데이터의 최적 복원과 데이터의 군집화를 위해 Autoencoder Neural Network (AENN) 과 Gradient-based FCM (GBFCM)을 이용하였다. OCAEFCM 의 성능평가를 위해 IRIS 데이터와 금융기관에서 취득한 실제 데이터를 사용하였다 기존의 Optimal Completion Strategy FCM (OCSFCM)과 비교했을 때, 제안된 OCAEFCM 이 OCSFCM 보다 18％-20％의 성능 향상을 보여준다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2000년도 제15차 학술회의논문집
• /
• pp.88-88
• /
• 2000

As more and more critical commercial applications move on the Internet, providing highly available servers becomes increasingly important. One of the advantages of a clustered system is that it has hardware and software redundancy. High availability can be provided by detecting node or daemon failure and reconfiguring the system appropriately so that the workload can be taken over bi the remaining nodes in the cluster. This paper presents how to provide the guaranteeing high availability of clustering web server. The load balancer becomes a single failure point of the whole system. In order to prevent the failure of the load balancer, we setup a backup server using heartbeat, fake, mon, and checkpointing fault-tolerance method. For high availability of file servers in the cluster, we setup coda file system. Coda is a advanced network fault-tolerance distributed file system.

• 한국정보과학회:학술대회논문집
• /
• 한국정보과학회 2003년도 가을 학술발표논문집 Vol.30 No.2 (1)
• /
• pp.10-12
• /
• 2003

Recommender systems help consumers to find the useful products from the overloaded information. Researchers have developed content-based recommenders, collaborative recommenders, and a few hybrid systems. In this research, we extend the classic collaborative recommenders by clustering method to form a hybrid recommender system. Using the clustering method, we can recommend the products based on not only the user ratings but also other useful information from user profiles or attributes of items. Through our experiments on well-known MovieLens data set, we found that the information provided by the attributes of item on the item-based collaborative filter shows advantage over the information provided by user profiles on the user-based collaborative filter.

• 한국컴퓨터정보학회논문지
• /
• 제24권6호
• /
• pp.89-97
• /
• 2019

This study proposes a method for investigating current patents related to information communication technology and smart mobility to provide insights into future technology trends. The method is based on text mining clustering analysis. The method consists of two stages, which are data preparation and clustering analysis, respectively. In the first stage, tokenizing, filtering, stemming, and feature selection are implemented to transform the data into a usable format (structured data) and to extract useful information for the next stage. In the second stage, the structured data is partitioned into groups. The K-medoids algorithm is selected over the K-means algorithm for this analysis owing to its advantages in dealing with noise and outliers. The results of the analysis indicate that most current patents focus mainly on smart connectivity and smart guide systems, which play a major role in the development of smart mobility.

• 제어로봇시스템학회논문지
• /
• 제19권1호
• /
• pp.27-33
• /
• 2013

Nowadays, the utilization of renewable energy sources like wind energy is considered one of the most effective means of generating massive amounts of electricity. This is evident in the rapid increase of wind farms all over the world which comprise a huge number of wind turbines. However, the drawback of utilizing wind turbines is that it requires maintenance, which could be a costly operation. To keep the wind turbines in pristine condition so as to reduce downtime, the implementation of CMS (Condition Monitoring System) and FDS (Fault Detection System) is mandatory. The efficiency and accuracy of these systems are crucial in deciding when to carry out a maintenance process. In this paper, a fault detection system based on intelligent clustering method is proposed. Using SCADA data, the clustering model was trained and evaluated for its accuracy through rigorous simulations. Results show that the proposed approach is able to accurately detect the deteriorating condition of a wind turbine as it nears a downtime period.

• ETRI Journal
• /
• 제37권2호
• /
• pp.283-294
• /
• 2015

A model that precisely forecasts how much wind power is generated is critical for making decisions on power generation and infrastructure updates. Existing studies have estimated wind power from wind speed using forecasting models such as ANFIS, SMO, k-NN, and ANN. This study applies a projected clustering technique to identify wind power patterns of wind turbines; profiles the resulting characteristics; and defines hourly and daily power patterns using wind power data collected over a year-long period. A wind power pattern prediction stage uses a time interval feature that is essential for producing representative patterns through a projected clustering technique along with the existing temperature and wind direction from the classifier input. During this stage, this feature is applied to the wind speed, which is the most significant input of a forecasting model. As the test results show, nine hourly power patterns and seven daily power patterns are produced with respect to the Korean wind turbines used in this study. As a result of forecasting the hourly and daily power patterns using the temperature, wind direction, and time interval features for the wind speed, the ANFIS and SMO models show an excellent performance.

• Journal of Information Processing Systems
• /
• 제13권3호
• /
• pp.435-447
• /
• 2017

Associative and bidirectional associative memories are examples of associative structures studied intensively in the literature. The underlying idea is to realize associative mapping so that the recall processes (one-directional and bidirectional ones) are realized with minimal recall errors. Associative and fuzzy associative memories have been studied in numerous areas yielding efficient applications for image recall and enhancements and fuzzy controllers, which can be regarded as one-directional associative memories. In this study, we revisit and augment the concept of associative memories by offering some new design insights where the corresponding mappings are realized on the basis of a related collection of landmarks (prototypes) over which an associative mapping becomes spanned. In light of the bidirectional character of mappings, we have developed an augmentation of the existing fuzzy clustering (fuzzy c-means, FCM) in the form of a so-called collaborative fuzzy clustering. Here, an interaction in the formation of prototypes is optimized so that the bidirectional recall errors can be minimized. Furthermore, we generalized the mapping into its granular version in which numeric prototypes that are formed through the clustering process are made granular so that the quality of the recall can be quantified. We propose several scenarios in which the allocation of information granularity is aimed at the optimization of the characteristics of recalled results (information granules) that are quantified in terms of coverage and specificity. We also introduce various architectural augmentations of the associative structures.

• International Journal of Control, Automation, and Systems
• /
• 제4권4호
• /
• pp.438-447
• /
• 2006

Fuzzy controller's design depends mainly on the rule base and membership functions over the controller's input and output ranges. This paper presents two different approaches to deal with these design issues. A simple and efficient approach; namely, Fuzzy Subtractive Clustering is used to identify the rule base needed to realize Fuzzy PI and PD type controllers. This technique provides a mechanism to obtain the reduced rule set covering the whole input/output space as well as membership functions for each input variable. But it is found that some membership functions projected from different clusters have high degree of similarity. The number of membership functions of each input variable is then reduced using a similarity measure. In this paper, the fuzzy subtractive clustering approach is shown to reduce 49 rules to 8 rules and number of membership functions to 4 and 6 for input variables (error and change in error) maintaining almost the same level of performance. Simulation on a wide range of linear and nonlinear processes is carried out and results are compared with fuzzy PI and PD type controllers without clustering in terms of several performance measures such as peak overshoot, settling time, rise time, integral absolute error (IAE) and integral-of-time multiplied absolute error (ITAE) and in each case the proposed schemes shows an identical performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제11권4호
• /
• pp.1866-1888
• /
• 2017

Efficient energy consumption in WSN is one of the key design issues for improving network stability period. In this paper, we propose a new Heterogeneity-aware Energy-efficient Clustering (HEC) technique which considers two types of heterogeneity - network lifetime and of sensor nodes. Selection of cluster head nodes is done based on the three network lifetime phases: only advanced nodes are allowed to become cluster heads in the initial phase; in the second active phase all nodes are allowed to participate in cluster head selection process with equal probability, and in the last dying out phase, clustering is relaxed by allowing direct transmission. Simulation-based performance analysis of the proposed technique as compared to other relevant techniques shows that HEC achieves longer stable region, improved throughput, and better energy dissipation owing to judicious consumption of additional energy of advanced nodes. On an average, the improvement observed for stability period over LEACH, SEP, FAIR and HEC- with SEP protocols is around 65%, 30%, 15% and 17% respectively. Further, the scalability of proposed technique is tested by varying the field size and number of sensing nodes. The results obtained are found to be quite optimistic. The impact of energy heterogeneity has also been assessed and it is found to improve the stability period though only upto a certain extent.

• 응용통계연구
• /
• 제24권4호
• /
• pp.623-631
• /
• 2011

시간경로 마이크로어레이 자료 분석의 주요 목적 중의 하나는 유전자들의 시간에 따른 발현수준의 변화를 고려함으로써 발현패턴에 기초한 유전자들의 그룹을 찾기 위한 것으로, 군집분석을 위한 다양한 알고리즘들이 제안되었다. 본 연구에서 시간경로 마이크로어레이 자료에 대한 군집분석을 위해 두 약물제제 간 생물학적 동등성을 평가하기 위한 약동학 시험에서 사용되는 약동학적 파라미터 값에 기초한 군집분석을 제안하였으며 이를 실제 데이터 및 모의실험 자료에 적용하여 유용성을 검토하였다.