• Journal of the Korean Institute of Intelligent Systems
• /
• v.16 no.6
• /
• pp.777-784
• /
• 2006

This paper proposes a c-fuzzy model tree using partial least square method to predict the Chlorophyll-a concentration in each zone. First, cluster centers are calculated by fuzzy clustering method using all input and output attributes. And then, each internal node is produced according to fuzzy membership values between centers and input attributes. Linear models are constructed by partial least square method considering input-output pairs remained in each internal node. The expansion of internal node is determined by comparing errors calculated in parent node with ones in child node, respectively. On the other hands, prediction is performed with a linear model haying the highest fuzzy membership value between input attributes and cluster centers in leaf nodes. To show the effectiveness of the proposed method, we have applied our method to water quality data set measured at several stations. Under various experiments, our proposed method shows better performance than conventional least square based model tree method.

• The Journal of the Korea Contents Association
• /
• v.9 no.1
• /
• pp.106-114
• /
• 2009

In this paper we proposed a fast method for a K-Means Clustering algorithm. The main characteristic of this method is that it uses precalculated data which possibility of change is high in order to speed up the algorithm. When calculating distance to cluster centre at each stage to assign nearest prototype in the clustering algorithm, it could reduce overall computation time by selecting only those data with possibility of change in cluster is high. Calculation time is reduced by using the distance information produced by K-Means algorithm when computing expected input data whose cluster may change, and by using such distance information the algorithm could be less affected by the number of dimensions. The proposed method was compared with original K-Means method - Lloyd's and the improved method KMHybrid. We show that our proposed method significantly outperforms in computation speed than Lloyd's and KMHybrid when using large size data which has large amount of data, great many dimensions and large number of clusters.

• The Korean Journal of Applied Statistics
• /
• v.34 no.6
• /
• pp.905-922
• /
• 2021

Cluster analysis is one of unsupervised learning techniques used for discovering clusters when there is no prior knowledge of group membership. K-means, one of the commonly used cluster analysis techniques, may fail when the number of variables becomes large. In such high-dimensional cases, it is common to perform tandem analysis, K-means cluster analysis after reducing the number of variables using dimension reduction methods. However, there is no guarantee that the reduced dimension reveals the cluster structure properly. Principal component analysis may mask the structure of clusters, especially when there are large variances for variables that are not related to cluster structure. To overcome this, techniques that perform dimension reduction and cluster analysis simultaneously have been suggested. This study proposes probabilistic reduced K-means, the transition of reduced K-means (De Soete and Caroll, 1994) into a probabilistic framework. Simulation shows that the proposed method performs better than tandem clustering or clustering without any dimension reduction. When the number of the variables is larger than the number of samples in each cluster, probabilistic reduced K-means show better formation of clusters than non-probabilistic reduced K-means. In the application to a real data set, it revealed similar or better cluster structure compared to other methods.

• Journal of Convergence for Information Technology
• /
• v.10 no.12
• /
• pp.126-131
• /
• 2020

This study analyzed the inflow and outflow patterns of emergency department patients, to identify changes in regional types in cities, counties, and districts in Jeonlado, Korea. Data of areas in Jeonlado for 2014 to 2018 were extracted from the National Emergency Department Information System. The extracted data includes the patients' and emergency medical institution addresses, which were used to calculate the relevance index (RI) and commitment index (CI). The calculated indices were classified into regional types by applying cluster analysis. A non-parametric method, Kruskal-Wallis test, was employed to examine the differences between years for RI and CI by regional types. The results of cluster analysis using the relevance and commitment indices revealed three regional types. Regions in cluster 1 were classified as outflow type, in cluster 2 as inflow type, and in cluster 3 as self-sufficient. RI and CI were calculated for each cluster or regional type. There were no significant differences between years in cluster 2 (inflow type) and cluster 3 (self-sufficient type). In cluster 1 (outflow type), there were no significant differences in CI between the years; however, there were significant differences in RI between 2014 and 2017, and 2014 and 2018. It is difficult to see that the emergency medical environment has improved due to the increased concentration of emergency medical care.

• Journal of Environmental Science International
• /
• v.12 no.12
• /
• pp.1215-1225
• /
• 2003

The purpose of this research was to investigate the characteristics of urban ecotopes and to classify ecotopes systematically from them. Total of 15 characteristics for classification of ecotopes were selected, and there were categorized 3 factors, that is abiotic, biotic and anthropological factors. The ecotope types in the study area were classified into 67. The classification of ecotope was made with SPSS for Windows Version 10.0 on the basis of the 15 characteristics. As the results of cluster analysis using the average linkage method between groups, groups of ecotope type were divided into 15 clusters. It was known that there was not a great difference in an affinity as the result of overlapping the maps of ecotope type and land use type. This research suggested characteristics for classification of ecotopes, but there was a limit to Set the objective method for grade classification because of lacking in the basic data, the research of characteristics will be accomplished continuously.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.12
• /
• pp.2230-2234
• /
• 2006

A SCO is a learning resource that is retrieved by a learner in the SCORM. A storage policy is required a learner to search SCOs rapidly in e-learning environment. In this paper, We define the mathematical formulation of clustering method for SCOs. Also we present criteria for cluster evaluation and describe procedure to evaluate each SCO. We show the search based on proposed clustering method increase performance than the existing search though performance evaluation.

• Journal of Korean Society for Quality Management
• /
• v.24 no.1
• /
• pp.126-136
• /
• 1996

In multi-item production system, GT(Group Technology) is used effectively in order to cluster various parts into groups. GT is based on clustering parts which have similar features, and these features are classified into two properties, namely crisp(quantitative) feature and fuzzy(qualitative) feature. Especially, many difficult problems are often faced that have to evaluate the properties of parts with the crisp and fuzzy feature together. As the basis of determining the similarity of inter-parts, in this method, one aggregate value is calculated on each part. However, because the above aggregate value is only gained from simple additive weighted sum, there is one problem in this method that has been handled the combination effect of inter-parts. For these reasons, in this paper, a proposed method is suggested for representing combination effect in order to cluster parts that have crisp and fuzzy properties into groups using ${\lambda}$-fuzzy measure and fuzzy integral.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.56 no.5
• /
• pp.876-883
• /
• 2007

The magnetic flux leakage(MFL) type nondestructive testing(NDT) method is widely used to detect corrosion, defects and mechanical deformation of the underground gas pipelines. The object pipeline is magnetically saturated by the magnetic system with permanent magnet and yokes. Hall sensors detect the leakage fields in the region of the defect. The defects are sometimes occurred in group. The accuracy of the detecting signals in this defect cluster become lowered because of the complexity of the defect cluster. In this paper, the effects of the multi -defects are analyzed. The detecting signals are computed by 3-dimensional finite element method and compared with real measurement. The results say that, rather than the size of the defects, the effects of the relative position of the multi-defects are very important on the detecting signals.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.1
• /
• pp.175-184
• /
• 2007

LOD scores and a permutation test for detecting and locating quantitative trait loci (QTL) from the Hanwoo economic trait have been described and we selected a considerable major BMS1167 locus for further analysis. K-means clustering analysis, for the major DNA marker mining of BMS1167 microsatellite loci in Hanwoo chromosome17, has been tried and three cluster groups divide four traits. The three cluster groups are classified according to eight DNA marker bps. Finally, we employed the bootstrap test method to calculate confidence intervals using the resampling method to find major DNA markers. We conclude that the major marker of BMS1167 locus in Hanwoo chromosome17 is only DNA marker 100bp.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.7 no.4
• /
• pp.31-37
• /
• 2011

Recently, MANET(Mobile Ad-hoc Network) is developing increasingly in the wireless network. MANET has weakness because phases change frequently and MANET doesn't have middle management system. Every node which consists of MANET has to perform data forwarding, but traceback is not reliable if these nodes do malicious action owing to attack. It also is not easy to find location of attacker when it is attacked as moving of nodes. In this paper, we propose a hierarchical-based traceback method that reduce waste of memory and can manage path information efficiently. In order to manage trace path information and reduce using resource in the cluster head after network is formed to cluster, method which recomposes the path efficiently is proposed. Proposed method in this paper can reduce path trace failure rate remarkably due to moving of nodes. It can also reduce the cost for traceback and time it takes to collect information.