• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.6 no.3
• /
• pp.210-216
• /
• 2006

In the rule based modeling, data partitioning plays crucial role be cause partitioned sub data set implies particular information of the given data set or system. In this paper, we present an empirical study result of the data pattern estimation to find underlying data patterns of the given data. Presented method performs crisp type clustering with given n number of data samples by means of the sequential agglomerative hierarchical nested model (SAHN). In each sequence, the average value of the sum of all inter-distance between centroid and data point. In the sequel, compute the derivation of the weighted average distance to observe a pattern distribution. For the final step, after overall clustering process is completed, weighted average distance value is applied to estimate range of the number of clusters in given dataset. The proposed estimation method and its result are considered with the use of FCM demo data set in MATLAB fuzzy logic toolbox and Box and Jenkins's gas furnace data.

• Journal of IKEEE
• /
• v.5 no.1 s.8
• /
• pp.100-110
• /
• 2001

In this paper, we investigated several buffer management schemes for the design of shared-memory type ATM switches, which can enhance the utilization of switch resources and can support quality-of-service (QoS) functionalities. Our results show that dynamic threshold (DT) scheme demonstrate a moderate degree of robustness close to pushout(PO) scheme, which is known to be impractical in the perspective of hardware implementation, under various traffic conditions such as traffic loads, burstyness of incoming traffic, and load non-uniformity across output ports. Next, we considered buffer management strategies to support QoS functions, which utilize parameter values obtained via connection admission control (CAC) procedures to set tile threshold values. Through simulations, we showed that the buffer management schemes adopted behave well in the sense that they can protect regulated traffic from unregulated cell traffic in allocating buffer space. In particular, it was observed that dynamic partitioning is superior in terms of QoS support than virtual partitioning.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.11 no.3
• /
• pp.253-257
• /
• 2013

Geology, hydrogeology, and geochemistry are the main technical siting factors of a geological repository for spent nuclear fuels. This paper evaluated the siting criteria of overseas geological repository with related to hydrogeologic properties, such as hydraulic conductivity, partitioning coefficient, dispersion coefficient, boundary condition, and water age. Each country establishes the siting criteria based on its important geological backgrounds and information, and social environment. For example, Sweden and Finland that have decided a crystalline rock as a host rock of a geological repository show different siting criteria for hydraulic conductivity. In Sweden, it is preferable to avoid area where the hydraulic conductivity on a deposition hole scale (~30m) exceeds $10^{-8}m/s$, whereas Finland does not decide any criterion for the hydraulic conductivity because of limited data for it. In addition, partitioning coefficients should be less than 10-1 of average value in Swedish crystalline bedrock. However, the area where shows 100 times less than average partitioning coefficients of radionuclides in crystalline rock should be avoided in Sweden. In German, the partitioning coefficients for the majority of the long-term-relevant radionuclides should be greater than or equal to $0.001m^3/kg$. Therefore, it is strongly required to collect much and exact information for the hydrogeologic properties in order to set up the siting criteria.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.21 no.8
• /
• pp.2003-2012
• /
• 1996

The 16- and 32-state Trellis-coded M-ary 4-dimensional (4-D) orthogonal modulation scheme with a frequency-reuse technique have been investigated. Here, 5 coded bits form a rate 4/5 convolutional encoder provide 32 possible symbols. Then the signals are mapped by a M-ary 4-D orthogonal modulator, where each signal has equal energy and is PSK modulated. In the M-ary 4-D modulator, we have employed the vectors which is derived by the optimization technique of signal waveforms in a 4-D sphere. This technique is usedin maximizing the minimum Euclidean distance between a set of signal poits on a multidimensional sphere. By combinig trellis coding with M-ary 4-D modulation and proper set-partitioning, we have obtained a considerable impeovement in the free minimum distance of the system over an AWGN channel. The 16-state scheme obtains coding gains up to 5.5 dB over the uncoded two-independent QPSK scheme and 2.5 dB over the two-independent 2-D TCM scheme. And, the 32-state scheme obtains coding gains up to 6.4 dB over the uncoded two-independent QPSK schemeand 3.4 dB over the two-independent 2-D TCM scheme.

• Journal of Institute of Control, Robotics and Systems
• /
• v.11 no.2
• /
• pp.127-136
• /
• 2005

In this paper, a new architecture and comprehensive design methodology of genetically optimized Rule-based Fuzzy Polynomial Neural Networks(gRFPNN) are introduced and a series of numeric experiments are carried out. The architecture of the resulting gRFPNN results from asynergistic usage of the hybrid system generated by combining rule-based Fuzzy Neural Networks(FNN) with polynomial neural networks (PNN). FNN contributes to the formation of the premise part of the overall rule-based structure of the gRFPNN. The consequence part of the gRFPNN is designed using PNNs. At the premise part of the gRFPNN, FNN exploits fuzzy set based approach designed by using space partitioning in terms of individual variables and comes in two fuzzy inference forms: simplified and linear. As the consequence part of the gRFPNN, the development of the genetically optimized PNN dwells on two general optimization mechanism: the structural optimization is realized via GAs whereas in case of the parametric optimization we proceed with a standard least square method-based learning. To evaluate the performance of the gRFPNN, the models are experimented with the use of several representative numerical examples. A comparative analysis shows that the proposed gRFPNN are models with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• Korean Journal of Remote Sensing
• /
• v.1 no.1
• /
• pp.5-27
• /
• 1985

A hill-sliding technique was devised to extract Gaussian clusters from the multivariate probability density estimates of sample data for the first step of iterative unsupervised classification. The underlying assumption in this approach was that each cluster possessed a unimodal normal distribution. The key idea was that a clustering function proposed could distinguish elements of a cluster under formation from the rest in the feature space. Initial clusters were extracted one by one according to the hill-sliding tactics. A dimensionless cluster compactness parameter was proposed as a universal measure of cluster goodness and used satisfactorily in test runs with Landsat multispectral scanner (MSS) data. The normalized divergence, defined by the cluster divergence divided by the entropy of the entire sample data, was utilized as a general separability measure between clusters. An overall clustering objective function was set forth in terms of cluster covariance matrices, from which the cluster compactness measure could be deduced. Minimal improvement of initial data partitioning was evaluated by this objective function in eliminating scattered sparse data points. The hill-sliding clustering technique developed herein has the potential applicability to decomposition of any multivariate mixture distribution into a number of unimodal distributions when an appropriate diatribution function to the data set is employed.

• Journal of Korea Multimedia Society
• /
• v.6 no.7
• /
• pp.1256-1265
• /
• 2003

Content Delivery Network(CDN) is a mechanism to deliver multimedia content to end users on behalf of web content providers. Provider's content is distributed from content server to a set of delivery platforms located at Internet Service Providers(ISPs) through the CDN in order to realize better performance and availability than the system of centralized provider's servers. Existing work on CDN has primarily focused on techniques for efficiently multicasting the content from content server to the delivery platforms or to end users. Multimedia contents usually require broader bandwidth and accordingly WDM broadcast network has been highly recommended for the infrastructure network of CDN. In this paper, we propose methods for partitioning a multicast group into smaller subgroups using the previous status of receivers. Through the computer simulation, we show that proposed algorithm are useful to reduce the average receiver's waiting time and the number of transmissions.

• Journal of KIISE:Software and Applications
• /
• v.29 no.3
• /
• pp.156-166
• /
• 2002

Decision tree induction is a kind of useful machine learning approach for extracting classification rules from a set of feature-based examples. According to the partitioning style of the feature space, decision trees are categorized into univariate decision trees and multivariate decision trees. Due to observation error, uncertainty, subjective judgment, and so on, real-world data are prone to contain some errors in their feature values. For the purpose of making decision trees robust against such errors, there have been various trials to incorporate fuzzy techniques into decision tree construction. Several researches hove been done on incorporating fuzzy techniques into univariate decision trees. However, for multivariate decision trees, few research has been done in the line of such study. This paper proposes a fuzzy decision tree induction method that builds fuzzy multivariate decision trees named fuzzy oblique decision trees, To show the effectiveness of the proposed method, it also presents some experimental results.

• Journal of Korea Multimedia Society
• /
• v.16 no.5
• /
• pp.657-666
• /
• 2013

Today, many aspects of our lives are characterized by the rapid influx of large amounts of data from various application domains. The applications that produce this massive of data span a large spectrum, from social media to business intelligence or biology. This massive influx of data necessitates large scale parallelism for efficiently supporting a large class of analysis tasks. Recently, there have been extensive studies in using MapReduce framework to support large parallelism. While this technique has produced impressive results in diverse applications, the same can not be said for multimedia applications where most of users are interested in a small number of results having high or low score. Thus, in this paper, we develop the data partitioning algorithm which is able to efficiently process large data set having different data utility. The experiment results show that the proposed technique provides significant execution time gains over the existing solution.

• Journal of KIISE:Databases
• /
• v.32 no.6
• /
• pp.616-627
• /
• 2005

In the shared storage file system, systems can directly access the shared storage device through specialized data-only subnetwork unlike in the network attached file server system. In this shared-storage architecture, data consistency is maintained by some designated set of lock servers which use control network to send and receive the lock information. Furthermore, lease mechanism is introduced to cope with the control network failure. But when the control network is partitioned, participating systems can no longer make progress after the lease term expires until the network recovers. This paper addresses this limitation and proposes a method that allows partitioned systems to make progress under the partition of control network. The proposed method works in a manner that each participating system is rotationally given a predefined lease term periodically. It is also shown that the proposed mechanism always preserves data consistency.