• The KIPS Transactions:PartB
• /
• v.13B no.4 s.107
• /
• pp.401-408
• /
• 2006

In this paper, we propose a new class of segmentation technique for feature extraction based on the statistical and regional classification at each vertical or horizontal line of digital image data. Data is processed and clustered at each line, different from the point or space process. They are designed to segment gray-scale sectional images using a horizontal and vertical line process due to their statistical and property differences, and to extract the feature. The techniques presented here show efficient results in case of the gray level overlap and not having threshold image. Such images are also not easy to be segmented by the global or local threshold methods. Line pixels inform us the sectionable data, and can be set according to cluster quality due to the differences of histogram and statistical data. The total segmentation on line clusters can be obtained by adaptive extension onto the horizontal axis. Each processed region has its own pixel value, resulting in feature extraction. The advantage and effectiveness of the line-cluster approach are both shown theoretically and demonstrated through the region-segmental carotid artery medical image processing.

• The Journal of Korean Medicine
• /
• v.24 no.1
• /
• pp.122-132
• /
• 2003

Objectives: To effectively improve the treatment of obesity through oriental medicine and to prepare basic material for proper classification of different types of obesity. Methods: After deciding on four types of obesity based on 'DongYiBaoJian', a questionnaire consisting of 38 items was constructed to decide to which type an individual belonged. 212 women were asked to the complete the questionnaire. To verify that the cluster of four types of obesity was acceptable, a cluster analysis and a factor analysis were conducted as well as an evaluation on the distinction of each type. Also, a canonical discriminant analysis was done to categorize the individuals into one of four types of obesity. Results: 1. Developed a reliable questionnaire consisting of 38 items for the purpose of classifying four types of obesity. 2. Obesity types were divided into four groups. Type I was designated as GanChengPi (肝乘脾類型), Type II as PiWeiJuWang(脾胃俱旺類型), Type III as PiWeiJuXu (脾胃俱虛類型), and Type IV as Tan TanYin(痰飮類型). These types were verified and classified through the use of a cluster analysis as well as a factor analysis (p<0.05). 3. By the use of a questionnaire, four types of obesity were correctly classified with a hit ratio of 87.3%, 40.64% higher than the maximum chance criteria (Cmax) in unselected grouped. The hit ratios for obesity types I, II, III and IV were 93.3%, 93.3%, 78.6% and 50%(p<0.05). Conclusion: Further clinical research is necessary into the four types of obesity explored. By analyzing various test results, characteristics these types should be further explored.

• Journal of the Korean Society of Costume
• /
• v.28
• /
• pp.99-112
• /
• 1996

In this study I purpose a new approach to the study of Korean costume history which includes a statistical method used in the prac-tical measurement of excavated Po's and the analysis of their quantitative changes. In the framework used in this study I de-pend on the nineteen variables picked out from the typological and chronological analysis of excavated clothes. The interrelationship of the variables as well as the standard of each part and its deviation in other forms are may main concern. Other methods are also very useful in my thesis: Correlation analysis Regression analysis Discrimination analysis Cluster analysis Discrimination analysis Cluster analysis Factor analysis One way analysis of variance. The findings of my study can be summarized as follows: 1. Every type of po has a general tendency to have more broadened sleeves while it does not show any considerable changes in the total length and in the width of armholes. 2. In the Correlation analysis each part of Po's is handled separately accordign to its chronology and type. The Regreesion analysis makes it possible to provide the sizes of miss-ing parts in excavated clothes and the numeri-cal value of each part needed in the pattern making. 3. The Discrimination analysis of the Po's whose types are unknown leads to the con-clusion that Chongsam and Chikryong Po are classified into Chikryong Po whereas Hongui is classified into Jooui In addition I propose discriminational function by which the age classification is possible discovering the variables associated with time change. 4. Five items-Gerneral Park's clothes Taewongun's clothes and three others produced later for the purpose of preserving Korean costume are proved to be in a differ-ent group and hence analyzed separately in the visual grouping of changui which is done after the cluster analysis and the factor analysis. 5. The results of the one way analysis of variance reveal that there is a difference in each part from period: the back width the sleeve length the sleeve width thelean fig-ure the breasttie length the breasttie width outside quesset upper inside quesset lower outside neckband length insideneckband length etc.

• Journal of Korea Multimedia Society
• /
• v.20 no.12
• /
• pp.1960-1969
• /
• 2017

There have been ongoing researches to identify and analyze the patterns of electric power IoT data inside sensor nodes to supplement the stable supply of power and the efficiency of energy consumption. This study set out to propose an analysis process for electric power IoT data with the K-means algorithm, which is an unsupervised learning technique rather than a supervised one. There are a couple of problems with the old K-means algorithm, and one of them is the selection of cluster number K in a heuristic or random method. That approach is proper for the age of standardized data. The investigator proposed an analysis process of selecting an automated cluster number K through principal component analysis and the space division of normal distribution and incorporated it into electric power IoT data. The performance evaluation results show that it recorded a higher level of performance than the old algorithm in the cluster classification and analysis of pitches and rolls included in the communication bodies of utility poles.

• Journal of Korean Society for Atmospheric Environment
• /
• v.33 no.6
• /
• pp.593-604
• /
• 2017

The suspended particulate matter(PM) was measured in subway tunnel of Seoul Line 1 to 9 in order to evaluate the pollution degree and characteristics of the PM in the subway tunnel. Also, to analyze the effect of outdoor aerosol concentration on the PM concentration of subway tunnels, the ambient PM concentration around the subway station was extracted by spatial analysis using $PM_{10}$ data of Seoul air pollution monitoring network. Finally, in order to understand pollution pattern in the Seoul subway tunnels, cluster analysis was performed based on input data set such as PM levels in tunnel, tunnel depth, length, curvature radius, outdoor ambient air pollution levels and so on. The average concentration of $PM_{10}$, $PM_{2.5}$, and $PM_1$ on subway tunnels were $98.0{\pm}37.4$, $78.4{\pm}28.7$, and $56.9{\pm}19.2{\mu}g/m^3$, respectively. As a result of the cluster analysis, tunnels from Seoul subway Line-1 to Line-9 were classified into five classes, and the concentrations and physical properties of the tunnels were compared. This study can provide a method to reduce PM concentration in tunnel for each pollution pattern and provide basic information about air quality control in Seoul subway tunnel.

• KIISE Transactions on Computing Practices
• /
• v.22 no.11
• /
• pp.601-613
• /
• 2016

With the increasing need for big data processing, building a shared big data processing platform is important to minimize time and monetary costs. In shared big data processing, multitenancy is a major requirement that needs to be addressed, in order to provide a single isolated personal big data platform for each user, but to share the underlying hardware is shared among users to increase hardware utilization. In this paper, we explore two well-known shared big data processing platform models. One is to use a native Hadoop cluster, and the other is to build a virtual Hadoop cluster for each user. For each model we verified whether it is sufficient to support multi-tenancy. We also present a method to complement unsupported multi-tenancy features in a native Hadoop cluster model. Lastly we built prototype platforms and compared the performance of both models.

• Journal of KIISE:Computing Practices and Letters
• /
• v.8 no.2
• /
• pp.183-194
• /
• 2002

The clustering environment with heterogeneous workstations provides the cost effectiveness and usability for executing applications in parallel. Load balancing is considered a necessary feature for a cluster of heterogeneous workstations to minimize the turnaround time. Previously, static load balancing that assigns a predetermined weight for the processing capability of each workstation, or dynamic approaches which execute a benchmark program to get relative processing capability of each workstation were proposed. The execution of the benchmark program, which has nothing to do with the application being executed, consumes the computation time and the overall turnaround time is delayed. In this paper, we present efficient methods for task distribution and task migration, based on the relative load index. We designed and implemented a load balancing system for the clustering environment with heterogeneous workstations. Turnaround times of our methods and the round-robin approach, as well as the load balancing method using a benchmark program, were compared. The experimental results show that our methods outperform all the other methods that we compared.

• Journal of Nutrition and Health
• /
• v.39 no.5
• /
• pp.460-466
• /
• 2006

To investigate the effects of dietary patterns on bone mineral density and its biochemical markers among Korean healthy college women for 2 years, 34 female college students were recruited through convenience sampling. Bone mineral density was measured using Dual Energy X-ray Absorptiometry (DEXA) twice at baseline and two years later. Osteocalcin and parathyroid hormone were measured in fasting serum and N-teleopeptides of type collagen (NTx) in urine. Dietary intake was assessed by 24-hour recall method 8 times with average 4-month interval. Dietary patterns with percent energy of each food group using cluster analysis were classified into two groups. The first cluster (n = 16) was characterized with high consumption of bread, snack, fast foods, beverage and considerable of rice so it was determined as 'Modified dietary pattern group'. The second cluster was characterized with high consumption of rice and kimchi so determined as 'Traditional dietary pattern group'. There were no significant difference of age, menarcheal age, body mass index but percent of body fat by pattern groups. The traditional group showed higher value of bone mineral density among lumber spine and all femur sites at baseline and 2 years later but it was not significant after adjusted for percent of body fat. Serum osteocalcin and urine NTx was higher among the traditional group at baseline than the modified group. There were similar proportions of carbohydrate:fat:protein between groups but significantly higher intake of protein, iron, vitamin A among the traditional group. In conclusion, there were two distinctive dietary patterns among Korean college women. There was difference of bone mineral density and its biochemical markers between two patterns. Further research would be necessary to explore the relationship between dietary patterns and health risks for larger-sized and various populations.

• Korean Journal of Medicinal Crop Science
• /
• v.2 no.2
• /
• pp.140-145
• /
• 1994

B. falcatum plants were classified into six groups from group I to grop VI by the complete linkage cluster method depending on 8 charactenstics such as plant height. number of nodes, number of branches, position of the first branching node root diameter, root length, number of lateral root, dry weight of root. These groups are divided into two plants types, such as multi-branching and non multi-branching type by the number of branches, group II and group VI were the multi-branching types and the other groups were nonmulti-branching ones, Dry weight of root had highly positive correlation with the number of branches and negative correlation with the position of first branching nodes.

• Journal of the Korean Crystal Growth and Crystal Technology
• /
• v.7 no.2
• /
• pp.207-215
• /
• 1997

Aluminum (Al) thin films were deposited on the Si(100) and TiN(60 nm)/Si (100) substrate by the ionized cluster beam deposition (ICBD) method. The characteristics of thin films were examined by the $\alpha$-step, four-point-probe, Scanning Electron Spectroscopy (SEM), Auger Electron Spectroscopy (AES). The growth rate of the Al thin film increased and the resistivity decreased as the crucible temperature increased. At the crucible temperature $1800^{\circ}C$, the microstructure of Al thin film deposited was smooth and continuous the resistivity decreased as the acceleration voltage increased. Also, the minimum resistivity in Si(100) substrate and TiN(60 nm)/Si(100) substrate were 3.4 $\mu \Omega \textrm {cm}$, 3.6 $\mu \Omega \textrm {cm}$ at the acceleration voltage 4 kV and 2 kV respectively. From the AES spectrumt 14 wasn't detected any impurities In the Al thin film. Therefore the resistivity of Al thin film was affected by the microstructure of film.