• Journal of the Korean Society for information Management
• /
• v.25 no.1
• /
• pp.211-233
• /
• 2008

This study examines various weighting methods for improving the performance of automatic classification based on Rocchio algorithm on two collections(LISA, Reuters-21578). First, three factors for weighting are identified as document factor, document factor, category factor for each weighting schemes, the performance of each was investigated. Second, the performance of combined weighting methods between the single schemes were examined. As a result, for the single schemes based on each factor, category-factor-based schemes showed the best performance, document set-factor-based schemes the second, and document-factor-based schemes the worst. For the combined weighting schemes, the schemes(idf*cat) which combine document set factor with category factor show better performance than the combined schemes(tf*cat or ltf*cat) which combine document factor with category factor as well as the common schemes (tfidf or ltfidf) that combining document factor with document set factor. However, according to the results of comparing the single weighting schemes with combined weighting schemes in the view of the collections, while category-factor-based schemes(cat only) perform best on LISA, the combined schemes(idf*cat) which combine document set factor with category factor showed best performance on the Reuters-21578. Therefore for the practical application of the weighting methods, it needs careful consideration of the categories in a collection for automatic classification.

• International Journal of Contents
• /
• v.1 no.2
• /
• pp.26-30
• /
• 2005

According to the fast growth of information on the Internet, it is becoming increasingly difficult to find and organize useful information. To reduce information overload, it needs to exploit automatic text classification for handling enormous documents. Support Vector Machine (SVM) is a model that is calculated as a weighted sum of kernel function outputs. This paper describes a document classifier for web documents in the fields of Information Technology and uses SVM to learn a model, which is constructed from the training sets and its representative terms. The basic idea is to exploit the representative terms meaning distribution in coherent thematic texts of each category by simple statistics methods. Vector-space model is applied to represent documents in the categories by using feature selection scheme based on TFiDF. We apply a category factor which represents effects in category of any term to the feature selection. Experiments show the results of categorization and the correlation of vector length.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.35 no.12
• /
• pp.1454-1465
• /
• 2011

Menswear body types are often labeled on garments (to indicate how the garments are designed to fit) with indicators of a size category such as regular, portly, and stout, athletic, or big and tall. A drop (relationships between the chest and waist girths) is related to the fit of a tailored suit. However, current standards are not designed for various drops or body types. There is not enough information of categorizing men's body shapes for the apparel sizing systems. In this article, a set of men's data from SizeUSA sizing survey was analyzed to investigate men's body shapes and drops. Factor analysis and a cluster analysis method were used to categorize men's body shapes. In the results, twenty-five variables were selected through the factor analysis and found four factors: girth factor, height factor, torso girth factor, and slope degree factor. According to the factor and cluster analysis, various body shapes were found: Slim Shape (SS - tall ectomorphy), Heavy Shape (HS - athletic, big & tall, endomorphy and mesomorphy), Slant Inverted Triangle Shape (SITS - regular, slight ectomorphy and slight mesomorphy weight range from normal to slightly overweight), Short Round Top Shape (SRTS - portly and stout, endomorphy). Body shapes were related to fitting categories. SS and HS were related to big & tall fitting category. SITS was related to regular. SRTS was related to portly and stout. Shape 1 (31%) and Shape 2 (26%) were related to current big & tall category. Shape 3 (34%) were related to regular. Shape 4 (9%) were in portly and stout category. ASTM D 6240 standard was the only available standard that presented a regular fitting category. Various drops were found within a same chest size group; however, this study revealed great variances of drops by body shape.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.16 no.4 s.44
• /
• pp.371-377
• /
• 1992

The purposes of the study were 1) to extend the cognitive categorization theory in an attempt to explain the of garment category, fashionability, and wearer's body types on impression formation, and 2) to find out structures of wearer's impressional dimension and wearer's professional image. The research included a quasi-experiment and survey. The experimental design was a $2^{3}$full factorial design of 3 independent variables. The experimental materials developed for the study were a set of stimuli and a response scale. The stimuli consisted of 8 drawings made by 3 independent variables (garment category, fashion level, wearer's body type). Result were as follows: 1) Garment category, fashionability and wearer's body type had significant effects on impression of the 5 factors-evaluation, potency, appearance, sociability and good-bad, with exception of wearer's body type which was nonsignificant to the potency factor. 2) Garment category was most effective on the evaluation and the potency. However wearer's body type was most effect on the appearance factor and fashionability variable was most effective on the good-bad factor. It was conclued that the results supported the cognitive categorization theory on impression formation and a cognitive categorization hypothesis of clothes.

• Journal of Information Processing Systems
• /
• v.16 no.4
• /
• pp.784-794
• /
• 2020

In order to effectively evaluate the urban water security, the study investigates a novel system to assess factors that impact urban water security and builds an urban water poverty evaluation index system. Based on the contribution rates of Resource, Access, Capacity, Use, and Environment, the study adopts the Water Poverty Index (WPI) model to evaluate the water poverty levels of 14 cities in Gansu during 2011-2018 and uses the least variance method to evaluate water poverty space drive types. The case study results show that the water poverty space drive types of 14 cites fall into four categories. The first category is the dual factor dominant type driven by environment and resources, which includes Lanzhou, Qingyang, Jiuquan, and Jiayuguan. The second category is the three-factor dominant type driven by Access, Use, and Capability, which includes Longnan, Linxia, and Gannan. The third category is the four-factor dominant type driven by Resource, Access, Capability, and Environment, which includes Jinchang, Pingliang, Wuwei, Baiyin, and Zhangye. The fourth category is the five-factor dominant type, which includes Tianshui and Dingxi. The driven types impacting the urban water security factors reflected by the WPI and its model are clear and accurate. The divisions of the urban water security level supply a reliable theoretical and numerical basis for an urban water security early warning mechanism.

• The Korean Journal of Health Service Management
• /
• v.5 no.2
• /
• pp.77-89
• /
• 2011

This study conducted a secondary analysis by using original data of performed by Korea Institute for Health and Social Affairs to know factors affecting determinants of health using Lalonde model for the adults aged over 19 years living in Korea. The survey was conducted in 2009 and it evaluated finally 5,867 cases by excluding cases with no answer or a wrong answer. This study model adopted two categories of instrument measure health were objective (Average remaining lifetime) and subjective(EQ-5D) health status. The health determinants included in this study could be divided in to four categories, which were human biology, environment, lifestyle, and health care organization. The results were as follows. In the factors affecting average remaining lifetime, human biology were sex, ages, BMI, showed statistically significant difference, environment category were merry status, education showed statistically significant difference, lifestyle category were exercise, drunks showed statistically significant difference and health care organization category were vaccination, health screening showed statistically significant difference. In the factors affecting EQ-5D, human biology category and health care organization category showed with same average remaining lifetime, environment category were merry status, education, income showed statistically significant difference and lifestyle category were exercise, drunks, stress showed statistically significant difference. The results demonstrated that the best powerful factor was life style category and environment category, the least factor was health care organization category. So lifestyle style and environment category should be considered for the future health plan, budget allocation and the priority in the health care.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.33 no.4
• /
• pp.1-12
• /
• 2008

The study suggests an answer to the question of what determines category diversification of Internet portals in Korea. First, as external factors, competition intensity and market growth are hypothesized to have influence on the degree of category diversification. Second, an internal factor, user loyalty to portals, is hypothesized to influence negatively category diversification. The study performed empirical analysis based on weekly portal-specific panel data of eighteen internet portals in Korea during the period between 2001 and 2004. The result shows that category diversification increases as competition intensity increases, and that category diversification decreases as user loyalty increases. There was no effect of market-level growth rate on category diversification.

• Knowledge Management Research
• /
• v.6 no.2
• /
• pp.105-118
• /
• 2005

The future will see all industries become technology-driven in the competitive global market place. Firms with deep technological roots and innovation strategies have some advantages. Business valuation of technology is critical to the future of firm's business. In this situation widely used scoring valuation is not enough to evaluate relative business competitiveness associated with technology and to assign its relative ranking category. Therefore, a more useful and comprehensive new valuation approach, which is called business composite index, is needed to complement and to enhance the existing scoring valuation approach. In this research, statistical factor analysis is applied to determine the common factors and to estimate associated weights. And business composite index, which is a kind of weighted scoring method, is derived based on the results of factor analysis. This research shows that business composite index is considered very useful to measure the business relative strength of individual technology and also to assign its relative ranking category instead of absolute ranking based on scoring valuation approach.

• Korea Real Estate Review
• /
• v.27 no.4
• /
• pp.7-20
• /
• 2017

This study attempts to diagnose and categorize the characteristics of old industrial parks, and eventually link the results to the regeneration of industrial complexes. For this reason, we performed a factor analysis by utilizing 15 indices of 89 industrial parks, excluding 5 large equipment industry sites. The 15 indices were classified into 5 factors. Factor 1 can be described as a category of 'urbanization possibility' for the indices of building age, plot ratio of less than $1,650m^2$, and urbanization ratio of the surrounding area. Factor 2 can be described as a category of 'productive efficiency' for the indices of land productivity, amount of exports by land, employment productivity, and repair costs of industrial areas. Factor 3 can be described as a category of 'infrastructure amenity' for the indices of road ratio, plot ratio attached to the road, and parks and recreation ratio. Factor 4 can be described as a category of 'location potentiality' for the indices of land price, infrastructure age, and distance to the highway, while factor 5 can be described as a category of 'availability of supporting facilities' for the indices of parking lot ratio and supporting facility land ratio. By using these 5 factor scores, we were able to extract industrial parks included in the lower 25% of the factor score and searched for what kind of factor problem they have for each industrial park. Based on these results, this research will provide sufficient information on the decline of industrial parks with respect to their demerits. The results of this study show significant implications and contribute to the establishment of policies for regional competitiveness, as well as job creation, in the process of industrial regeneration.

• Nutrition Research and Practice
• /
• v.6 no.2
• /
• pp.162-168
• /
• 2012

Vitamin D deficiency is a risk factor for metabolic syndromes. We examined whether vitamin D deficiency altered the prevalence of cardiovascular disease (CVD) in older Koreans. Cross-sectional analysis of data from the Korean National Health and Nutrition Examination Survey IV 2008-2009 was used to examine the association between serum 25-hydroxyvitamin D (25(OH)D) levels and the prevalence of CVD in a representative population-based sample of 5,559 men and women aged ${\geq}50$ years. CVD was defined as angina pectoris, myocardial infarction, or stroke. The prevalence of CVD (7.0%) in the older Korean population was lower than that in the older US population, although average serum 25(OH)D levels were much lower in the Korean population. Additionally, serum 25(OH)D levels did not differ significantly between the CVD and non-CVD groups. However, subjects in the lowest category (< 25 nmol/l) of serum 25(OH)D level had the greatest prevalence of CVD, about two-fold higher than subjects in the highest category (> 75 nmol/l), after adjusting for age, gender, body mass index, education level, residence location, and region. The prevalence of other risk factors for CVD, including higher waist circumference, fasting glucose, low-density lipoprotein (LDL) cholesterol, and triglyceride levels and lower high-density lipoprotein (HDL) cholesterol levels, was also higher among subjects in the lowest category than among those in the highest category. In conclusion, low serum 25(OH)D may be an independent risk factor for CVD in older Koreans.