• Proceedings of the CALSEC Conference
• /
• 한국전자거래학회 2004년도 e-Biz World Conference
• /
• pp.294-301
• /
• 2004

·Text-mining approach to e-documents organization based on topic hierarchy - Machine-Learning ＆ information Theory-based ㆍ 'Category(topic) discovery' problem → document bundle-based user-constraint document clustering ㆍ 'Automatic categorization' problem → Accelerated EM with CU-based active learning → 'Hierarchy Construction' problem → Unsupervised learning of category subsumption relation

• Journal of KIISE:Software and Applications
• /
• 제37권5호
• /
• pp.378-385
• /
• 2010

The explosive increase in the use of email has made to need email classification efficiently and accurately. Current work on the email classification method have mainly been focused on a binary classification that filters out spam-mails. This methods are based on Support Vector Machines, Bayesian classifiers, rule-based classifiers. Such supervised methods, in the sense that the user is required to manually describe the rules and keyword list that is used to recognize the relevant email. Other unsupervised method using clustering techniques for the multi-category classification is created a category labels from a set of incoming messages. In this paper, we propose a new automatic email multi-category classification method using NMF for automatic category label construction method and dynamic category hierarchy method for the reorganization of email messages in the category labels. The proposed method in this paper, a large number of emails are managed efficiently by classifying multi-category email automatically, email messages in their category are reorganized for enhancing accuracy whenever users want to classify all their email messages.

• Journal of information and communication convergence engineering
• /
• 제7권3호
• /
• pp.351-355
• /
• 2009

The amount of incoming e-mails is increasing rapidly due to the wide usage of Internet. We often group e-mails into categories for maintaining e-mail efficiently. However reading the email messages and classifying them is still tedious task. Moreover, the number of e-mails and manual classifying is increasing everyday. So, automatic e-mail classification is important techniques. In this paper, we propose a multi-way e-mail classification method that uses PCA for automatic category generation and dynamic category hierarchy for re-organizing e-mail categories. It classifies a huge amount of receiving e-mail messages automatically, efficiently, and accurately.

• Journal of Korea Multimedia Society
• /
• 제21권12호
• /
• pp.1417-1424
• /
• 2018

In order to improve the performance of image classifications using Convolutional Neural Networks (CNN), applying a category hierarchy to the classification can be a useful idea. However, the visual separation of object categories is very different according to the upper and lower category levels and highly uneven in image classifications. Therefore, it is doubtable whether the use of category hierarchies for classification is effective in CNN. In this paper, we have clarified whether the image classification using category hierarchies improves classification performance, and found at which level of hierarchy classification is more effective. For experiments we divided the image classification task according to the upper and lower category levels and assigned image data to each CNN model. We identified and compared the results of three classification models and analyzed them. Through the experiments, we could confirm that classification effectiveness was not improved by reduction of number of categories in a classification model. And we found that only with the re-training method in the last network layer, the performance of lower category classification was not improved although that of higher category classification was improved.

• Journal of KIISE:Software and Applications
• /
• 제37권3호
• /
• pp.214-221
• /
• 2010

The search-results clustering problem is defined as the automatic and on-line grouping of similar documents in search results returned from a search engine. In this paper, we propose a new search-results clustering algorithm specialized for a paper search service. Our system consists of two algorithmic phases: Category Hierarchy Generation System (CHGS) and Paper Clustering System (PCS). In CHGS, we first build up the category hierarchy, called the Field Thesaurus, for each research field using an existing research category hierarchy (KOSEF's research category hierarchy) and the keyword expansion of the field thesaurus by a word clustering method using the K-means algorithm. Then, in PCS, the proposed algorithm determines the category of each paper using top-down and bottom-up methods. The proposed system can be used in the application areas for retrieval services in a specialized field such as a paper search service.

• Journal of Advanced Navigation Technology
• /
• 제13권3호
• /
• pp.419-425
• /
• 2009

The amount of incoming e-mails is increasing rapidly due to the wide usage of Internet. Therefore, it is more required to classify incoming e-mails efficiently and accurately. Currently, the e-mail classification techniques are focused on two way classification to filter spam mails from normal ones based mainly on Bayesian and Rule. The clustering method has been used for the multi-way classification of e-mails. But it has a disadvantage of low accuracy of classification and no category labels. The classification methods have a disadvantage of training and setting of category labels by user. In this paper, we propose a novel multi-way e-mail hierarchy classification method that uses PCA for automatic category generation and dynamic category hierarchy for high accuracy of classification. It classifies a huge amount of incoming e-mails automatically, efficiently, and accurately.

• Korean Journal of Cognitive Science
• /
• 제11권3_4호
• /
• pp.73-81
• /
• 2000

This study was conducted to investigate some characteristics of representation of category knowledge and script knowledge. Using primed lexical decision task with higher level primers in the representation structure, Experiment 1 examined the interaction effects between knowledge type and concept typicality. It was found that the concept typicality has some effects in category representation, while it has no significant effect in script representation. In Experiment 2, primers of the lower hierarchy in the representation structure were employed. The results showed that the main effect of knowledge type was significant: the response time for category knowledge was faster than that for script knowledge. Typicality effect did not show in this experiment. The results of t the two experiments suggest that category knowledge is represented in hierarchy and typicality. while script knowledge may lack in that characteristics. Other aspects of the differences in characteristics of category- and script- knowledge representation were discussed,

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 한국해양정보통신학회 2009년도 춘계학술대회
• /
• pp.576-579
• /
• 2009

The amount of incoming e-mails is increasing rapidly due to the wide usage of Internet. Therefore, it is more required to classify incoming e-mails efficiently and accurately. Currently, the e-mail classification techniques are focused on two way classification to filter spam mails from normal ones based mainly on Bayesian and Rule. The clustering method has been used for the multi-way classification of e-mails. But it has a disadvantage of low accuracy of classification. In this paper, we propose a novel multi-way e-mail classification method that uses PCA for automatic category generation and dynamic category hierarchy for high accuracy of classification. It classifies a huge amount of incoming e-mails automatically, efficiently, and accurately.

• Korea Journal of Hospital Management
• /
• 제6권3호
• /
• pp.148-166
• /
• 2001

The purpose of this study is to classify elderly patient in long-term care facilities using RUG(Resource Utilization Group)-III. It is designed by measuring patient medical characteristics and medical staff time. Elderly patients are classified into 7 categories by clinical(medical and behavioral) hierarchical typology of patients. Through the tertiary split, all 44 groups are formulated. This classification is explained by each patient resource(staff time) utilization level which is called CMI(Case-Mix Index). Major findings are as follows; 1. The objects in this study were classified into 35 groups out of 44 groups. The most frequent category is clinical complex category(CCC; 38.9%). And extensive service category(ESC; 18.8%), reduced physical function category(RPC; 13.1%), special rehabilitation category(SRC; 12.8%), and impaired cognitive category(ICC; 0.00%) are followed. 2. The mean of total CMI was $1.02{\pm}0.36$, ranging from 0.68 to 1.44(1 vs 2.12). The mean of CMI of SRC is only 1.17 which should be the highest. The means of ESC and see are equally 1.20. The means of CMI of CCI, ICC, BPC, and RPC were 0.90, 0.75, 0.83 and 0.96, respectively. 3. The validity of this classification was tested. Trend-test using Regression Analysis was done in the secondary split level. SCC, CCC, ICC, and RPC which covered 68.4% of this research objects showed linear trend of CMI in interim classification. This results were statistically significant. 4. In clinical hierarchy, the trend were showed linearity. But the multiple comparison of categories using Scheffe-test showed that SRC, ESC and see had same level of CMI means and CCC and ICC, too. This results were statistically significant. Classifying elderly patients with RUG-III, the results showed partly linear trend in clinical hierarchy and in interim classification in conclusion. But, in clinical hierarchy, it was failed to show the consistent order of CMI. It can be explained by two reasons. One is that this research subjects were overlapped in each clinical hierarchy group. And the other is that the some of the characteristics for clinical hierarchy is not appropriate for them. For the further study, it needs to have proper sample size and to modify RUG-III to K-RUG to consider our.. medical environment.

• Journal of Nutrition and Health
• /
• 제39권1호
• /
• pp.74-83
• /
• 2006

The purposes of this study were to (1) identify the evaluation categories, areas, attributes, and criteria of the school food service program using both a qualitative and a quantitative analyses, (2) define the relative importance of the evaluation categories, areas, attributes, and criteria of the school food service program using analytic hierarchy process, (3) organize the evaluation system to improve quality of the school food service in Korea. A survey was conducted from August to October 2004 to collect data from 172 dietitians, 15 school food service officials at the educational board, 10 professionals of school food service. Statistical analyses were performed on the data utilizing the SPSS 12.0 for Windows and Excel, such as Descriptive statistics and analytic hierarchy process was performed. The result of the analytic hierarchy process indicated that relative importance of evaluation category was 0.4319 (food service manage ment), 0.2369 (nutrition education), 0.1455 (satisfaction) and 0.0912 (parent involvement program). 'Sanitation, safety and facility (0.1739)' was the most important area among the subcategories of food service management, followed by nutrition management (0.1581), procurement (0.1375), production (0.1345), organization and personnel management (0.0662), planning (0.0644), food service evaluation (0.0585), financial accountability (0.0555), and information management (0.0554). There existed a relative importance on the three areas of the nutrition program and satisfaction evaluation category: students (0.5281, 0.6221), parents (0.1812, 0.1491), and teachers (0.1838, 0.1618). In the parent involvement program evaluation category, relative importance of committee and monitoring management was 0.4658 and information communication was 0.3724. The quality of food and service to school children can be improved by the appropriate application of the developed evaluation tool for the school food service program.