• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.1
• /
• pp.95-99
• /
• 2010

Social media refers to the content, which are created by users, such as blogs, social networks, and wikis. Recently, question-answering (QA) communities, in which users share information by questions and answers, are regarded as a kind of social media. Thus, QA communities have become a huge source of information for the past decade. However, it is hard for users to search the exact question-answer that is exactly matched with their needs as the number of question-answers increases in QA communities. This paper proposes an approach for classifying a question into three categories (information, opinion, and suggestion) according to the purpose of the question for more accurate information retrieval. Specifically, our approach is based on modified Na$\ddot{i}$ve Bayes classifier which uses structural characteristics of QA documents to improve the classification accuracy. Through our experiments, we achieved about 71.2% in classification accuracy.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.15 no.2
• /
• pp.37-52
• /
• 2005

Traditional access control models like role-based access control model are insufficient in security needs in ubiquitous computing environment because they take no thought of access control based on user's context or environment condition. In these days, although researches on context-aware access control using user's context or environment conditions based on role-based access control are emerged, they are on the primary stage. We present context definitions md an access control model to provide more flexible and dynamic context-aware access control based on role-based access control. Specially, we describe the conflict problems occurred in the middle of making an access decision. After classifying the conflict problems, we show some resolutions to solve them. In conclusion, we will lay the foundations of the development of security policy and model assuring right user of right object(or resource) and application service through pre-defined context and context classification in ubiquitous computing environments. Beyond the simplicity of access to objects by authorized users, we assure that user can access to the object, resource, or service anywhere and anytime according to right context.

• 한국HCI학회:학술대회논문집
• /
• 2008.02b
• /
• pp.193-198
• /
• 2008

Now days with the significant grows of internet and PC hardware/software, most of the business, include small size business and soho, is worked by PC/Network. However these critical changes, all the PC/Network system or service is identical and these make user a inconvenient experience and unpleasant aspect to PC/Network. This research is to segment the user bases on the user context of using PC and all this segmentation was done in contextual design process. From Contextual Interview, all data of user context of environment, use case, needs, voices, break downs were captured and these data were interpreted and filtered to affinity notes and user modeling. After the interpretation, the consolidation was being held by affinity diagram, which could make ready to segment the user. For the result, 3 segmentation was evaluated and which are the 'Holistic Networker', 'Unplugged Controller' and 'Passive Reactor'.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.7
• /
• pp.942-948
• /
• 2020

In this paper, we propose a clustered algorithm that possible more efficient user distinction within clustering using context-aware attribute information. In typically, the data provided to classify interrelationships within cluster information in the process of clustering data will be as a degrade factor if new or newly processing information is treated as contaminated information in comparative information. In this paper, we have developed a clustering algorithm that can extract user's recognition information to solve this problem in using K-means algorithm. The proposed algorithm analyzes the user's clustering attributed parameters from user clusters using accumulated information and clustering according to their attributes. The results of the simulation with the proposed algorithm showed that the user management system was more adaptable in terms of classifying and maintaining multiple users in clusters.

• The Journal of Society for e-Business Studies
• /
• v.21 no.1
• /
• pp.147-163
• /
• 2016

Gender information of a smart device user is essential to provide personalized services, and multi-modal data obtained from the device is useful for predicting the gender of the user. However, the method for utilizing each of the multi-modal data for gender prediction differs according to the characteristics of the data. Therefore, in this study, an ensemble method for predicting the gender of a smart device user by using three classifiers that have text, application, and acceleration data as inputs, respectively, is proposed. To alleviate privacy issues that occur when text data generated in a smart device are sent outside, a classification method which scans smart device text data only on the device and classifies the gender of the user by matching text data with predefined sets of word. An application based classifier assigns gender labels to executed applications and predicts gender of the user by comparing the label ratio. Acceleration data is used with Support Vector Machine to classify user gender. The proposed method was evaluated by using the actual smart device log data collected from an Android application. The experimental results showed that the proposed method outperformed the compared methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2000.10a
• /
• pp.305-308
• /
• 2000

웹사이트 설계자의 주관적 판단에 의한 정적 하이퍼텍스트 링킹은 모든 사용자들에게 동일한 링크를 제공한다는 단점을 가지고 있다. 이러한 문제점을 개선하고, 각 사용자들의 브라우징 패턴에 적합한 웹 문서들을 동적 링크로 제공해주기 위한 여러 동적 링킹 시스템들이 제안되었다. 그러나 대부분의 동적 링킹 시스템들은 사용자의 현재 브라우징 패턴과 가장 유사한 패턴 정보만을 이용해 동적 링크를 제공하기 때문에 연관성이 없는 웹 문서들에 대한 링크를 수시로 제공한다는 또 다른 문제를 지니고 있다. 본 논문에서는 데이터 마이닝의 한 응용 분야인 웹 마이닝 기법을 이용하여 웹 서버의 로그파일로부터 사용자들의 브라우징 패턴을 분석해내고, 다차원 데이터 집합에 적합한 Association Rule Hypergraph Partitioning(ARHP) 알고리즘을 이용하여 서로 연관성이 있는 웹 문서들을 분류한다. 사용자 브라우징 패턴 정보로부터 사용자에게 추천해줄 1차 링크 집합을 생성하고, 연관 웹 문서 정보를 이용하여 2차 링크 집합을 생성한다. 그리고 두 링크 집합에 공통으로 포함된 링크 집합만을 사용자에게 동적으로 추천해줌으로써 사용자가 보다 편리하고 정확하게 웹사이트를 브라우징 할 수 있도록 하는 동적 링킹 시스템을 제안한다.

• Journal of Digital Convergence
• /
• v.8 no.4
• /
• pp.177-184
• /
• 2010

Mobile phones users are not only able to use smart phone services in real time whenever and wherever they desire, but also they can acquire appropriate information that they need. The features of smart phone services and users are examined in this study because there are not enough studies on such unique features of smart phone services. In accordance with the results of the study the features of smart phone services and users can be categorized as follows: 1) the requisite features of smart phone services are immediate connectivity and. 2) The requisite features for smart phone users are user friendliness and user innovation. 3) The requisite features of smart phone technology are security and linkage.

• Journal of KIISE:Software and Applications
• /
• v.30 no.1_2
• /
• pp.51-57
• /
• 2003

For QA systems to return correct answer phrases, it is very important that they correctly and stably analyze users' intention. To satisfy this need, we propose a question type classifier (i.e. asking point identifier) for practical QA systems. The classifier uses a hybrid method that combines a statistical method with a rule-based method according to some heuristic rules. Owing to the hybrid method, the classifier can reduce the time to manually construct rules, yield high precision rate and guarantee robustness. In the experiment, we accomplished 80% accuracy of the question type classification.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.214-216
• /
• 2006

온라인 정보가 증가함에 따라 많은 양의 정보 중에서 사용자가 원하는 정보를 정확하고 신속하게 찾아 주는 문서 여과의 중요성 또한 증가하고 있는 추세이다. 본 논문은 문서 여과 문제를 이진 문서 분류 문제로 보고, 나이브 베이즈 분류기를 동적인 문서 여과 목적으로 사용하였다. 이때 사용자가 자신의 관심 분야에 해당하는 주제를 제대로 여과 받기 위해서 학습 대상으로 삼아야 할 학습문서의 범위와 관련성 있는 문서를 제대로 여과 받기 위해서 체크해야 하는 관련성 표기 비율에 따른 분류기의 성능에 대하여 실험을 하였다. 코사인 유사계수를 이용한 여과 방법과의 성능도 비교 실험하였다. 실험 결과 나이브 베이즈 이진 분류기는 문서집합의 크기가 일정한 정도일 때 관련성 있는 문서가 모두 표기되지 않더라도 여과에는 큰 영향을 미치지 않음을 볼 수 있었다.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2645-2647
• /
• 2003

E-Mail은 현재 가장 많이 쓰이고 있는 인터넷 서비스로서 최근 인터넷 사용자의 증가와 함께 그 사용자 또한 늘어나고 있다. 기존의 메일 기반 에이전트는 서버에 무분별하게 메일을 저장하는 방식이기 때문에, 첨부파일에 접근하기 위해서는 관리자가 수신된 메일을 일일이 확인해야 하는 번거로움을 가지고 있다. 대량의 메일을 수신하는 기업에서는 메일 뿐만 아니라 첨부되어오는 파일들에 대한 처리부담이 더욱 크다. 본 논문에서는 이를 보완하기 위해 도착한 메일의 내용을 분석하여 키워드를 추출하고, 폴더를 생성하여 카테고리별로 첨부된 파일을 분류해주는 자동분류 에이전트를 제안하고자 한다. 카테고리 별로 분류된 파일은 다시 형식별로 분류되도록 설계하였다. 이는 관리자의 업무부담을 줄이고, 첨부파일을 효과적으로 관리할 수 있는 장점이 있다.