• Annual Conference on Human and Language Technology
• /
• 2017.10a
• /
• pp.116-121
• /
• 2017

cQA(Community-based Question Answering) 시스템은 온라인 커뮤니티를 통해 사용자들이 질문을 남기고 답변을 작성할 수 있도록 만들어진 시스템이다. 신규 질문이 인입되면, 기존에 축적된 cQA 저장소에서 해당 질문과 가장 유사한 질문을 검색하고, 그 질문에 대한 답변을 신규 질문에 대한 답변으로 대체할 수 있다. 하지만, 키워드 매칭을 사용하는 전통적인 검색 방식으로는 문장에 내재된 의미들을 이용할 수 없다는 한계가 있다. 이를 극복하기 위해서는 의미적으로 동일한 문장들로 학습이 되어야 하지만, 이러한 데이터를 대량으로 확보하기에는 어려움이 있다. 본 논문에서는 질문이 제목과 내용으로 분리되어 있는 대량의 cQA 셋에서, 질문 제목과 내용을 의미 벡터 공간으로 사상하고 두 벡터의 상대적 거리가 가깝게 되도록 학습함으로써 의사(pseudo) 유사 의미의 성질을 내재화 하였다. 또한, 질문 제목과 내용의 의미 벡터 표현(representation)을 위하여, semi-training word embedding과 CNN(Convolutional Neural Network)을 이용한 딥러닝 기법을 제안하였다. 유사 질문 검색 실험 결과, 제안 모델을 이용한 검색이 키워드 매칭 기반 검색보다 좋은 성능을 보였다.

• The Journal of the Korea Contents Association
• /
• v.16 no.7
• /
• pp.423-430
• /
• 2016

The results of the response in the Self-report survey are occurred differently according to situational factors and personal characteristics. This study attempts to examine the effects of the question-order and question-intentions in surveys of job satisfaction. Four types of questionnaires were used in consideration of the difference between Whole Questions and detail Questions in question-order and the difference between job satisfaction measurement and job dissatisfaction measurement in question-intentions. This study surveyed 94 employees. As a result, the order and intention of questions could affect respondent's answers. However, It was not statistically significant support. Also, this study suggest significant implications, limits of this study and research direction of future study.

• Proceedings of the Korean Society for Information Management Conference
• /
• 1995.08a
• /
• pp.17-20
• /
• 1995

본 연구는 질의어 확장 및 단기 이용자 모형 구축에 응용할 수 있는 하나의 기법으로서 이용자 질문구조의 전형을 통한 정보요구의 모형화를 실험을 통해 제시한다. 실험방법은 이용자의 질문을 시소러스를 통해 분석, 구조화 하고 그 질문구조에서 전형을 추출한 후 전형에 따라 요구하는 정보가 질문구조내에 일정하게 위치하는지를 알아보았다. 이러한 실험을 통해 6가지 질문구조 전형을 추출할 수 있었으며 질문구조의 전형을 이용한 정보요구 모형의 구축이 타당성이 있음을 입증하였고 지능형 정보검색 시스템에의 적용가능성을 논의하였다.

• The Korean Journal of Applied Statistics
• /
• v.11 no.2
• /
• pp.415-421
• /
• 1998

In this paper, we proposed improved unrelated question model which has the benefit of simplicity the Kim et al.'s two-stage unrelated question model(1992). conditions are obtained under which the proposed model is more efficient than the Greenberg et al. model(1971) and Kim et al's two-stage unrelated question model.

• Journal of the Korean Chemical Society
• /
• v.55 no.1
• /
• pp.124-131
• /
• 2011

This study is conducted to analyze the questioning styles of the parts of chemistry in 17 middle school science textbooks in terms of frequency, types, and placement of questions. The instrument was the Textbook Questioning Strategies Assessment Instrument (TQSAI) developed by the Cooperative Teacher Preparation Program, University of California in 1975. An averaged ratio of questionings/sentences (Q/S) of seventeen textbooks was 13.6% varying from 6.8% to 19.7%. Various question types were used in titles, early and middle steps in the sections except late steps. The comprehending questions were most frequent in the types of inquiry courses and the focusing questions were most frequent in the characteristics of instructional courses. The percentages of empirical questions for the total numbers of questions were slightly larger than nonempirical questions with values of 54.2% and 45.8%, respectively.

• Journal of The Korean Association For Science Education
• /
• v.24 no.2
• /
• pp.277-286
• /
• 2004

This study investigated the types and frequencies of student-student questioning (SSQ) in a small group activities, 5 in one group or 2 in one group, during school experiments. Five girls of seventh grade were observed during school experiments and interviewed afterward. Between students, information-type questions were asked more frequently than thought-type questions. Most of the information-type questions were procedural ones and most of the thought-type questions were comprehension ones. However, thought-type questions did not make further discussion in the activities. The rate of answers in the case of 2 in one group was higher than that of 5 in one group. Moreover, the similar tendency was found when we investigated the rate of helpful question-answers. In a pair, lower achiever usually asked questions, not answered as much as in 5 in one group, and higher achiever answered. The frequency of SSQ in a pair was relatively low when there was a big difference of science achievements between pair members. In conclusion, information-type questions were asked more frequently than thought-type questions during school experiments and the rate of helpful question-answers was higher when group members was fewer.

• Science of Emotion and Sensibility
• /
• v.21 no.2
• /
• pp.113-124
• /
• 2018

The logical assumption of the comparison question test (CQT) is that the guilty person pays more attention to the relevant questions than to the comparison questions, and that the innocent person pays more attention to the comparison questions than to the relevant questions. The purpose of this study was to verify the logic of the comparison question test using functional magnetic resonance imaging (fMRI). The participants were tested for brain responses during a mock crime and performed the CQT under guilty and innocent conditions. After brain imaging, we evaluated the psychological burden of responding to the relevant questions and comparison questions. In the guilty conditions, the degree of burden was higher for the relevant questions than the comparison questions, and there was no significant difference in the innocent conditions. The fMRI results showed that, in the guilty conditions, greater activation was observed in the right superior temporal gyrus and right inferior frontal gyrus when relevant questions were presented relative to comparison questions. Based on these findings, the logical assumption of the CQT was discussed.

• 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 Korean Association For Science Education
• /
• v.19 no.4
• /
• pp.560-569
• /
• 1999

This study was conducted with two science classes of the 8th grade students in Seoul during 4 weeks. The numbers of students in the classes were 37 and 38, and they were taught for 12 class hours. Questions obtained for 12 class hours from 75 students were analyzed and grouped into patterns. All together 1.108 questions from the students were classified into six categories: 'No Connection' (7%). 'Contradiction' (3%), 'Recall' (23%), 'Reframe' (40%), 'Application' (18%), and 'Extension' (9%). Irrelevant questions to learning and questions of false proposition were classified into 'No Connection' and 'Contradiction', respectively. Questions repeating what were already explained were grouped into 'Recall'. Those requiring other examples and/or additional explanations were grouped into 'Reframe'. Those requiring practical applications and/or explanations for other concrete facts were grouped into 'Application'. Finally. Questions for higher and/or other concepts were grouped into 'Extension'. We also discussed educational implications of the above categorized questions in this study.

• Journal of the Korean Society for information Management
• /
• v.18 no.2
• /
• pp.125-142
• /
• 2001

This study attempts to identify users' potential information needs in a medium-sized academic library by analyzing reference questions raised by library users for 64 days from October 10th to December 13th, 2000. A total of 474 reference questions were analysed of which only 10.1% were pure reference questions that dealt with the professional knowledge of reference librarians. 71.0% had nothing to do with professional knowledge, being directional(29.3%), mechanic(23.8%), and library policy related(l7.9%). Improvements in sign system and guiding materials in the library should be made to deliver reference services of good quality to academic library users.