• IE interfaces
• /
• v.15 no.2
• /
• pp.182-188
• /
• 2002

Clustering is widely used in various fields in order to investigate structural characteristics of the given data. One of the main tasks of clustering is to partition a set of objects into homogeneous groups for the purpose of data reduction. In this paper a simple but computationally efficient clustering procedure is devised and some statistical techniques to validate its clustered results are discussed. In the given procedure, the proper number of clusters and the clustered groups can be determined simultaneously. The whole procedure is applied to a practical clustering problem for the classification of reusable fuels in nuclear power plants.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.18 no.4
• /
• pp.560-565
• /
• 1994

The purpose of this study was to extract information of body form's classification on elderly women. We measured 242 subjects from 55 to 75 years of age, using 27 direct measurement items and 25 body indices. We analyzed these indices with factor analysis, cluster analysis We obtained these following results 1) Through factor analysis, 4 factors (obesity of torso, location of upper torso items, length of upper torso, location of lower torso items & shoulder length) were extracted from body indices. 2) Through culster analysis, we categorized 4 clusters. Namely, type 1, characterized the best slender type, type 2; characterized obesity type, type 3: characterized middle sized type ; type 4: characterized by fat type less than type 2. We considered that type 3 is the typical type on elderly women. Since analysis using indices is very profitable, it may be necessary to design dummies and patterns for clothing manufacture.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.33 no.3
• /
• pp.93-105
• /
• 2008

Semi-supervised learning uses a small amount of labeled data to predict labels of unlabeled data as well as to improve clustering performance, whereas unsupervised learning analyzes only unlabeled data for clustering purpose. We propose a new clustering-based semi-supervised learning method by reflecting the initial predicted labels of unlabeled data on the objective function. The initial prediction should be done in terms of a discrete probability distribution through a classification method using labeled data. As a result, clusters are formed and labels of unlabeled data are predicted according to the Information of labeled data in the same cluster. We evaluate and compare the performance of the proposed method in terms of classification errors through numerical experiments with blinded labeled data.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.22 no.3
• /
• pp.333-340
• /
• 1998

The purpose of this study was to classify the lower body somatotype based on the side view and to analyze the characteristics of each somatotype. The subject were 234 Korean Junior High School Girls. Data were collected through photographic sources and analyzed by factor analysis, cluster analysis and analysis of variance. The result of factor analysis indicated that 4 factors were extracted through factor analysis and those factorscomprised 73.5% of total variance. Using factor scores, cluster analysis was carried out and the subject were classified into 3 clusters. Each cluster was classified as their lower bobs side view contour.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.29 no.5 s.142
• /
• pp.737-744
• /
• 2005

This study was aimed to provide the measurement data and shape classification of the head of the Korean male children for the headwear sizing systems. Five hundred twenty male children, aged nine to twelve years, participated f3r this study. The 17 regions on the head and height, weight of the subjects were directly measured by the expert experimenters. Factor analysis, cluster analysis, GLM analysis and Tukey HSD test were performed using these data. Through factor analysis, low factors were extracted upon factor scores and those factors comprised $69.76\%$ for the total variances. Three clusters as their head shape were categorized using four factor scores by cluster analysis. Type 1 was characterized by the widest width and Bitragion arc, shortest head length. Type 2 had the longest head length and the widest side width and the highest head length and head circumference. Type 3 was characterized by the smallest head circumstance, head width and side width, and medium head length.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.47 no.7
• /
• pp.13-22
• /
• 2005

Main purpose of this article is to classify reservoirs corresponding to their physical characteristics, for example, dam height, dam width, age, repair-works history. First of all, data set of 13,976 reservoirs was analyzed using k means and self organized maps. As a result of these analysis, lots of reservoirs have been classified into four clusters. Factors and their critical values to classify the reservoirs into four groups have been founded by generating a decision tree. The path rules to each group seem reasonable since their survivor function showed unique pattern.

• Communications for Statistical Applications and Methods
• /
• v.17 no.3
• /
• pp.377-390
• /
• 2010

In the case of time-series analysis, it is often more convenient to rely on the frequency domain than the time domain. Spectral density is the core of the frequency-domain analysis that describes autocorrelation structures in a time-series process. Possible ways to estimate spectral density are to compute a periodogram or to average the periodogram over some frequencies with (un)equal weights. This can be an attractive tool to measure the similarity between time-series processes. We employ the metrics based on a smoothed periodogram proposed by Park and Kim (2008) for the classification of different classes of time-series processes. We consider several lag windows with unequal weights instead of a modified Daniel's window used in Park and Kim (2008). We evaluate the performance under various simulation scenarios. Simulation results reveal that the metrics used in this study split the time series into the preassigned clusters better than do the raw-periodogram based ones proposed by Caiado et al. 2006. Our metrics are applied to an economic time-series dataset.

• Korean Journal of Remote Sensing
• /
• v.20 no.3
• /
• pp.215-225
• /
• 2004

A multistage hierarchical clustering technique, which is an unsupervised technique, was suggested in this paper for classifying large remotely-sensed imagery. The multistage algorithm consists of two stages. The 'local' segmentor of the first stage performs region-growing segmentation by employing the hierarchical clustering procedure of CN-chain with the restriction that pixels in a cluster must be spatially contiguous. The 'global' segmentor of the second stage, which has not spatial constraints for merging, clusters the segments resulting from the previous stage, using the conventional agglomerative approach. Using simulation data, the proposed method was compared with another hierarchical clustering technique based on 'mutual closest neighbor.' The experimental results show that the new approach proposed in this study considerably increases in computational efficiency for larger images with a low number of bands. The technique was then applied to classify the land-cover types using the remotely-sensed data acquired from the Korean peninsula.

• Proceedings of the KSRS Conference
• /
• 2004.10a
• /
• pp.428-431
• /
• 2004

A multi-stage hierarchical clustering technique, which is an unsupervised technique, has been proposed in this paper for classifying the hyperspectral data .. The multistage algorithm consists of two stages. The 'local' segmentor of the first stage performs region-growing segmentation by employing the hierarchical clustering procedure with the restriction that pixels in a cluster must be spatially contiguous. The 'global' segmentor of the second stage, which has not spatial constraints for merging, clusters the segments resulting from the previous stage, using a context-free similarity measure. This study applied the multistage hierarchical clustering method to the data generated by band reduction, band selection and data compression. The classification results were compared with them using full bands.

• Korean Journal of Remote Sensing
• /
• v.19 no.2
• /
• pp.170-179
• /
• 2003

In order to classify Land Use/Land Cover using multispectral images, we have to give consequence to defining proper classes and selecting training sample with higher class separability. The process of satellite hyperspectral image which has a lot of bands is difficult and time-consuming. Furthermore, classification result of hyperspectral image with noise is often worse than that of a multispectral image. When selecting training fields according to the signatures in the study area, it is difficult to calculate covariance matrix in some clusters with pixels less than the number of bands. Therefore in this paper we presented an overview of feature extraction methods for classification of Hyperion data and examined effectiveness of feature extraction through the accuracy assesment of classified image. Also we evaluated the classification accuracy of optimal meaningful features by class separation distance, which is also a method for band reduction. As a result, the classification accuracies of feature-extracted image and original image are similar regardless of classifiers. But the number of bands used and computing time were reduced. The classifiers such as MLC, SAM and ECHO were used.