• Journal of information and communication convergence engineering
• /
• v.6 no.4
• /
• pp.434-438
• /
• 2008

Massive data are continuously produced with a data rate of over several terabytes every day. These applications need effective clustering algorithms to achieve an overall high performance computation. In this paper, we propose ancestor as cluster center based approach to clustering, the K-means algorithm using ancestor. We modify the K-means algorithm. We present a clustering architecture and a clustering algorithm that minimize of I/Os and show a performance with excellent. In our experimental performance evaluation, we present that our algorithm can improve the I/O speed and the query processing time.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.671-672
• /
• 2008

K-means is a popular one in clustering algorithms, and it minimizes the mutual euclidean distance among the sample points. But K-means has some demerits, such as depending on initial condition, unsupervised learning and local optimum. However mahalanobis distancecan deal this case well. In this paper, the author proposed a new clustering algorithm, named exponential probability clustering, which applied Mahalanobis distance into K-means clustering. This new clustering does possess not only the probability interpretation, but also clustering merits. Finally, the simulation results also demonstrate its good performance compared to K-means algorithm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.2
• /
• pp.136-141
• /
• 2004

There has been enormous growth in the amount of commercial and scientific data, such as retail transactions, protein sequences, and web-logs. Such datasets consist of sequence data that have an inherent sequential nature. However, few clustering algorithms consider sequentiality. In this paper, we study how to cluster sequence datasets. We propose a new similarity measure to compute the similarity between two sequences. We also present an efficient method for determining the similarity measure and develop a clustering algorithm. Due to the high computational complexity of hierarchical clustering algorithms for clustering large datasets, a new clustering method is required. Therefore, we propose a new scalable clustering method using sampling and a k-nearest-neighbor method. Using a real dataset and a synthetic dataset, we show that the quality of clusters generated by our proposed approach is better than that of clusters produced by traditional algorithms.

• Journal of Korea Society of Industrial Information Systems
• /
• v.16 no.3
• /
• pp.47-56
• /
• 2011

To apply Genetic algorithm toward document clustering, appropriate individual structure is required. Document clustering with the genetic algorithms (DCGA) uses the centroid vector type individual structure. New document clustering with the genetic algorithm (NDAGA) uses document allocated individual structure. In this paper, to find more suitable object structure and process for the document clustering, calculation, amount of calculation, run-time, and performance difference between the two methods were analyzed. In this paper, we have performed various experiments using both DCGA and NDCGA. Result of the experiment shows that compared to DCGA, NDCGA provided 15% faster execution time, about 5~10% better performance. This proves that the document allocated structure is more fitted than the centroid vector type structure when it comes to document clustering. In addition, NDCGA showed 15~25% better performance than the traditional clustering algorithms (K-means, Group Average).

• The KIPS Transactions:PartC
• /
• v.16C no.6
• /
• pp.783-790
• /
• 2009

The necessity of wireless sensor networks is increasing in the recent years. So many researches are studied in wireless sensor networks. The clustering algorithm provides an effective way to prolong the lifetime of the wireless sensor networks. The one-hop routing of LEACH algorithm is an inefficient way in the energy consumption of cluster-head, because it transmits a data to the BS(Base Station) with one-hop. On the other hand, other clustering algorithms transmit data to the BS with multi-hop, because the multi-hop transmission is an effective way. But the multi-hop routing of other clustering algorithms which transmits data to BS with multi-hop have a data bottleneck state problem. The unequal clustering algorithm solved a data bottleneck state problem by increasing the routing path. Most of the unequal clustering algorithms partition the nodes into clusters of unequal size, and clusters closer to the BS have small-size the those farther away from the BS. However, the energy consumption of cluster-head in unequal clustering algorithm is more increased than other clustering algorithms. In the thesis, I propose an energy efficient unequal clustering algorithm which decreases the energy consumption of cluster-head and solves the data bottleneck state problem. The basic idea is divided a three part. First of all I provide that the election of appropriate cluster-head. Next, I offer that the decision of cluster-size which consider the distance from the BS, the energy state of node and the number of neighborhood node. Finally, I provide that the election of assistant node which the transmit function substituted for cluster-head. As a result, the energy consumption of cluster-head is minimized, and the energy consumption of total network is minimized.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.4
• /
• pp.721-727
• /
• 2016

In this paper, in order to resolve the problem of reduction for D2D (device to device) advertisement dissemination efficiency of conventional dissemination algorithms, we here propose several clustering algorithms (modified single linkage algorithm (MSL), K-means algorithm, and expectation maximization algorithm with Gaussian mixture model (EM)) based advertisement dissemination algorithms to improve advertisement dissemination efficiency in D2D communication networks. Target areas are clustered in several target groups by the proposed clustering algorithms. Then, D2D advertisements are consecutively distributed by using a routing algorithm based on the geographical distribution of the target areas and a relay selection algorithm based on the distance between D2D sender and D2D receiver. Via intensive MATLAB simulations, we analyze the performance excellency of the proposed algorithms with respect to maximum number of relay transmissions and D2D user density ratio in a target area and a non-target area.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.26 no.1
• /
• pp.17-21
• /
• 2003

We study clustering algorithm for sequences of categorical values. Clustering is a data mining problem that has received significant attention by the database community. Traditional clustering algorithms deal with numerical or categorical data points. However, there exist many important databases that store categorical data sequences. In this paper, we introduce new similarity measure and develop a hierarchical clustering algorithm. An experimental section shows performance of the proposed approach.

• Journal of Information Processing Systems
• /
• v.2 no.2
• /
• pp.67-75
• /
• 2006

In order to acquire a precise and quick response to an analytical query, proper selection of the views to materialize in the data warehouse is crucial. In traditional view selection algorithms, all relations are considered for selection as materialized views. However, materializing all relations rather than a part results in much worse performance in terms of time and space costs. Therefore, we present an improved algorithm for selection of views to materialize using the clustering method to overcome the problem resulting from conventional view selection algorithms. In the presented algorithm, ASVMRT (Algorithm for Selection of Views to Materialize using Reduced Table), we first generate reduced tables in the data warehouse using clustering based on attribute-values density, and then we consider the combination of reduced tables as materialized views instead of a combination of the original base relations. For the justification of the proposed algorithm, we reveal the experimental results in which both time and space costs are approximately 1.8 times better than conventional algorithms.

• Proceedings of the KIEE Conference
• /
• 2005.05a
• /
• pp.9-11
• /
• 2005

We propose a optimal identification of information granulation(IG)-based fuzzy model to carry out the model identification of complex and nonlinear systems. To optimally identity we use genetic algorithm (GAs) sand Hard C-Means (HCM) clustering. An initial structure of fuzzy model is identified by determining the number of input, the selected input variables, the number of membership function, and the conclusion inference type by means of GAs. Granulation of information data with the aid of Hard C-Means(HCM) clustering algorithm help determine the initial parameters of fuzzy model such as the initial apexes of the membership functions and the initial values of polynomial functions being used in the premise and consequence part of the fuzzy rules. And the initial parameters are tuned effectively with the aid of the genetic algorithms(GAs) and the least square method. Numerical example is included to evaluate the performance of the proposed model.

• Journal of Electrical Engineering and information Science
• /
• v.2 no.6
• /
• 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.