• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2008.10b
• /
• pp.164-169
• /
• 2008

The increasing late-payment rate of credit card customers caused by a recent economic downturn are incurring not only reduced profit of department stores but also significant loss. Under this pressure, the objective of credit forecasting is extended from presumption of good or bad customers to contribution to revenue growth. As a method of managing defaults of department store credit card, this study classifies credit delinquents into some clusters, analyzes repaying patterns of customers in each cluster, and develops credit forecasting system to manage delinquents of department store credit card using data of Korean D department store's delinquents. The model presented by this study uses Kohonen network, a kind of artificial neural network of data mining techniques to cluster credit delinquents into groups. Logistic regression model is also used to predict repayment rate of customers of each cluster per period. The accuracy of presented system for the whole clusters is 92.3%.

• Korean Journal of Plant Resources
• /
• v.19 no.6
• /
• pp.716-720
• /
• 2006

The genetic diversity among the genus Viola was evaluated using the random amplified polymorphic DNA (RAPD) method. A total of 142 distinct amplification fragments by 18 random primers were scored to perform the cluster analysis with UPGMA. Viola species from the subsection Patellares were clustered into group I to IV. The groups from I to IV were consistent with its morphological taxonomy, series Pinnatae, Chinensis, Variegatae, and Patellares in the subsection Patellares, respectively. Even though V. albida and V. albida var. takahasii were classified in Chinensis, they were assigned into group I. The cluster analysis separated other subsections from Patellares in the section Nomimium. Interestingly, V. verecunda and V. grypoceras in subsections Biobatae and Trigonocarpae, respectively, were clustered into group C with a high similarity coefficient. Therefore, RAPD analysis can be used for providing an alternative classification system to identify genotypes and morphological characters of Viola species.

• Journal of the Korea Safety Management & Science
• /
• v.11 no.3
• /
• pp.157-166
• /
• 2009

In this study, we developed a heuristic algorithm to get better efficiency of clustering than conventional algorithms. Conventional clustering algorithm had lower efficiency of clustering as there were no solid method for selecting initial center of cluster and as they had difficulty in search solution for clustering. EMC(Expanded Moving Center) heuristic algorithm was suggested to clear the problem of low efficiency in clustering. We developed algorithm to select initial center of cluster and search solution systematically in clustering. Experiments of clustering are performed to evaluate performance of EMC heuristic algorithm. Squared-error of EMC heuristic algorithm showed better performance for real case study and improved greatly with increase of cluster number than the other ones.

• The Bulletin of The Korean Astronomical Society
• /
• v.41 no.1
• /
• pp.71.2-71.2
• /
• 2016

Merging galaxy clusters, such as PLCKG287.0+32.9, provide a window into the formation process of the large scale structure of the universe. PLCKG287.0+32.9 is an enormous merging galaxy cluster with mass estimated to be ~10^15 Msun. It hosts a pair of mega-parsec sized radio relics with projected offsets from the X-ray center of approximately 350kpc and 2.7Mpc, suggesting a NW-SE merging scenario with relics originating from two separate passes (Bonafede et al. 2014). A detected radio halo coincides with the center of x-ray emission. We present the motivation for our weak lensing study of the merging galaxy cluster PLCKG287.0+32.9 using recent Subaru optical imaging. We discuss the basics of weak-lensing and the criteria for source selection. In addition, we describe our method of PSF modeling and mass reconstruction.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.15 no.2 s.38
• /
• pp.151-162
• /
• 1991

The purpose of this study was to examine the point at issue unfitness and unclassification in figures for clothing construction. Comparing measurements of 3 sold figures and physical body by Suck·Go method and 19 items were selected. These items were acquired by factor analysis and cluster analysis an measurements of 200 women in twenties. The results were as follows; 1. Comparison between measurement of figures and physical body; There was not so much difference among figures, but significant difference between figures and physical body. 2. Comparison between measurement of surface area of figures and physical body; There was similar to each figure, but significant difference between figures and physical body. Mostly, figures were bigger than physical body in front parts but smaller in back parts. 3. There were 8 items selected by factor analysis; That is bust girth, waist girth, hip girth, neck base girth, posterior waist length, bust point length, chest breadth and armscye girth. 4. There were 7 clusters selected by cluster analysis.

• Journal of the Korean Institute of Electrical and Electronic Material Engineers
• /
• v.20 no.11
• /
• pp.959-964
• /
• 2007

To investigate the influence of partial discharge(PD) distribution characteristics due to various defects on the power cable joints interface, we used the K-means clustering method. As the result of PD number(n) distribution analyzing on $\Phi-n$ graph, the phase angle($\Phi$) of cluster centroid shifted to $0^{\circ}\;and\;180^{\circ}$ increasing with applying voltage. It was confirmed that the PD quantify(q) and euclidean distance of centroid were increased with applying voltage from the centroid distribution analyzing of $\Phi-q$ plane. The dispersion degree was increased with calculated standard deviation of the $\Phi-q$ cluster centroid. The PD number and mean value on $\Phi-q$ graph were some different by electric field concentration with defect types.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2013.08a
• /
• pp.123.2-123.2
• /
• 2013

Etching endpoint detection with plasma impedance monitoring (PIM) is demonstrated for small area dielectric layers inductive coupled plasma etching. The endpoint is determined by the impedance harmonic signals variation from the I-V monitoring system. Measuring plasma impedance has been examined as a relatively simple method of detecting variations in plasma and surface conditions without contamination at low cost. Cluster analysis algorithm is modified and applied to real-time endpoint detection for sensitivity enhancement in this work. For verification, the detected endpoint by PIM and real-time cluster analysis is compared with widely used optical emission spectroscopy (OES) signals. The proposed technique shows clear improvement of sensitivity with significant noise reduction when it is compared with OES signals. This technique is expected to be applied to various plasma monitoring applications including fault detections as well as end point detection.

• Journal of Veterinary Science
• /
• v.22 no.6
• /
• pp.88.1-88.6
• /
• 2021

Toxoplasma gondii consists of three genotypes, namely genotype I, II and III. Based on its virulence, T. gondii can be divided into virulent and avirulent strains. This study intends to evaluate an alternative method for predicting T. gondii virulence using hierarchical cluster analysis based on complete coding sequences (CDS) of sag1, gra7 and rop18 genes. Dendrogram was constructed using UPGMA with a Kimura 80 nucleotide distance measurement. The results showed that the prediction errors of T. gondii virulence using sag1, gra7 and rop18 were 7.41%, 6.89% and 9.1%, respectively. Analysis based on CDS of gra7 and rop18 was able to differentiate avirulent strains into genotypes II and III, whereas sag1 failed to differentiate.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.1
• /
• pp.157-163
• /
• 2023

Recently, as businesses and data types become more complex and diverse, efficient data analysis using machine learning is required. However, since communication in the cloud environment is greatly affected by network latency, data analysis is not smooth if information delay occurs. In this paper, SPT (Safe Proper Time) was applied to the cluster-based machine learning data analysis agent proposed in previous studies to solve this delay problem. SPT is a method of remotely and directly accessing memory to a cluster that processes data between layers, effectively improving data transfer speed and ensuring timeliness and reliability of data transfer.

• The Journal of the Acoustical Society of Korea
• /
• v.6 no.4
• /
• pp.31-43
• /
• 1987

As a study on Speaker-Independent Isolated Word Recognition, a Modified ISODATA clustering method is proposed. This method simplifies the outlier processing and the splitting procedure in conventional ISODATA algorithm, and eliminates the lumping procedure. Through this method, we could find cluster centers precisely and automatically. When this method applied to 11 digits by 10 males and 4 females, its recognition rates of $84.42\%$ for K=4 were better than those of the latest Modified K-means, $82.5\%$. Judging from these results, we proved this method the best method in finding cluster centers precisely.