• Journal of information and communication convergence engineering
• /
• v.7 no.4
• /
• pp.419-430
• /
• 2009

Reading comprehension (RC) systems are to understand a given text and return answers in response to questions about the text. Many previous studies extract sentences that are the most similar to questions as answers. However, texts for RC tests are generally short and facts about an event or entity are often expressed in multiple sentences. The answers for some questions might be indirectly presented in the sentences having few overlapping words with the questions. This paper proposes a conceptual graph matching method towards RC tests to extract answer strings. The method first represents the text and questions as conceptual graphs, and then extracts subgraphs for every candidate answer concept from the text graph. All candidate answer concepts will be scored and ranked according to the matching similarity between their sub-graphs and question graph. The top one will be returned as answer seed to form a concise answer string. Since the sub-graphs for candidate answer concepts are not restricted to only covering a single sentence, our approach improved the performance of answer extraction on the Remedia test data.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.12
• /
• pp.267-274
• /
• 2014

Even though computer based college lecture management system was developed long ago and has been used ever since, developing perfect lecture management is not simple. The main objective of this system is to develop appropriate online lecture supportive management program suitable for college lectures. This system [ADLEMS] mainly focuses on the management of college database lectures, exams, and grades. This system supports management and grading of attendance, reports, quizzes, mid-term, and final exams. Exam management categorizes into multiple choice questions, essay questions, short answer questions, and SQL. Especially for SQL, division analysis was applied when developing grading system for more effective grade management. For essay questions and short answer questions, manual [hand] grading method was used. Every student can verify the grading process in person to alleviate the problems occurring during the grading process.

• KIPS Transactions on Software and Data Engineering
• /
• v.5 no.11
• /
• pp.527-534
• /
• 2016

The emergent information society requires the talent for creative thinking based on problem-solving skills and comprehensive thinking rather than simple memorization. Therefore, the Korean curriculum has also changed into the direction of the creative thinking through increasing short-answer questions that can determine the overall thinking of the students. However, their scoring results are a little bit inconsistency because scoring short-answer questions depends on the subjective scoring of human raters. In order to alleviate this point, an automated scoring system using a machine learning has been used as a scoring tool in overseas. Linguistically, Korean and English is totally different in the structure of the sentences. Thus, the automated scoring system used in English cannot be applied to Korean. In this paper, we introduce an automated scoring system for Korean short-answer questions using predictability and unanimity. We also verify the practicality of the automatic scoring system through the correlation coefficient between the results of the automated scoring system and those of human raters. In the experiment of this paper, the proposed system is evaluated for constructed-response items of Korean language, social studies, and science in the National Assessment of Educational Achievement. The analysis was used Pearson correlation coefficients and Kappa coefficient. Results of the experiment had showed a strong positive correlation with all the correlation coefficients at 0.7 or higher. Thus, the scoring results of the proposed scoring system are similar to those of human raters. Therefore, the automated scoring system should be found to be useful as a scoring tool.

• The Journal of the Korea Contents Association
• /
• v.6 no.3
• /
• pp.12-19
• /
• 2006

In this paper, we describe a deep analysis of question for question answering system. It is difficult to offer the correct answer because general question answering systems do not analyze the semantic of user's natural language question. We analyze user's question semantically and extract semantic features using the semantic feature extraction grammar and characteristics of natural language question. They are represented as semantic features and grammatical morphemes that consider semantic and syntactic structure of user's questions. We evaluated our approach using 100 questions whose answer type is a person in the web. We showed that a deep analysis of questions which are comparatively short but enough to mean can analysis the user's intention and extract semantic features.

• Speech Sciences
• /
• v.13 no.3
• /
• pp.207-219
• /
• 2006

The present study was designed to investigate the comprehension of 'who' and 'what' questions in 2- to 3-year-old normal children. Sixty children were divided into 3 groups depending on their ages, i.e., age groups 2;6-2;11, 3;0-3:5, and 3;6-3;11. Three types of 'who' questions and 2 types of 'what' questions were generated depending on the attached case markers, i.e., who-nominative, who-accusative, who-dative, what-nominative, and what-accusative. The children watched 36 cuts of short video recordings. After watching each cut, they were asked to answer one of the 5 types of wh-questions. For the 'who-nominative' and 'what-accusative' questions, even the late 2-year-old children performed with over 70% of accuracy, and the late 3-year-old children performed with over 95% of accuracy. For the 'who-accusative' and 'who-dative' questions, the late 2-year olds exhibited difficulty in comprehension with performance accuracy of 41% and 33%, respectively. However, the late 3-year olds could comprehend those questions correctly with over 90% of accuracy. On the other hand, in answering 'what-nominative' questions, the children did not show rapid development across the age groups, as the mean performance accuracies of the 3 groups were 39%, 49%, and 59%, respectively. The results indicated that children's understanding of a wh- question is largely affected by the case of the interrogative.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.3
• /
• pp.135-141
• /
• 2012

Development of the Internet has brought many changes in methods of education and assesment. When enforcing the on-line distance education, the tests to check the outcomes of the learning are taken on the Internet. The current trends of education evaluation are focused on the types of questions and the detachments of exam proctor but verifying the authentication of answer sheet. There are several forms to make answers; selection type, short-answer type, write-out answer type, practical exercise type, etc. All the forms can be done on the Internet except the practical exercise type because the source of the examinee's answer sheet is unreliable. In this paper, we made the verification system to solve the doubt by setting the proved information on the answer sheet. Putting the information down to confirm the authenticity during the exam on the server is distinct character of this system. After the test finished, the system will operate when examinee turn in the answer sheet.

• The Mathematical Education
• /
• v.57 no.2
• /
• pp.137-155
• /
• 2018

From the assumption that an individual's working memory capacity is limited, the cognitive load theory is concerned with providing adequate instructional design so as to avoid overloading the learner's working memory. Based on the cognitive load theory, this study aimed to provide implications for effective problem-based collaborative teaching and learning design by analyzing the level of middle school students' cognitive load which is perceived according to the problem types(short answer type, narrative type, project) in the process of collaborative problem solving in middle school function part. To do this, this study analyzed whether there is a relevant difference in the level of cognitive load for the problem type according to the math achievement level and gender in the process of cooperative problem solving. As a result, there was a relevant difference in the task burden and task difficulty perceived according to the types of problems in both first and second graders in middle schools students. and there was no significant difference in the cognitive effort. In addition, the efficacy of task performance differed between first and second graders. The significance of this study is as follows: in the process of collaborative problem solving learning, which is most frequently used in school classrooms, it examined students' cognitive load according to problem types in various aspects of grade, achievement level, and gender.

• Journal of KIBIM
• /
• v.13 no.3
• /
• pp.21-29
• /
• 2023

ChatGPT, a chatbot based on GPT large language models, has gained immense popularity among the general public as well as domain professionals. To assess its proficiency in specialized fields, ChatGPT was tested on mainstream exams like the bar exam and medical licensing tests. This study evaluated ChatGPT's ability to answer questions related to Building Information Modeling (BIM) by testing it on Korea's BIM expertise exam, focusing primarily on multiple-choice problems. Both GPT-3.5 and GPT-4 were tested by prompting them to provide the correct answers to three years' worth of exams, totaling 150 questions. The results showed that both versions passed the test with average scores of 68 and 85, respectively. GPT-4 performed particularly well in categories related to 'BIM software' and 'Smart Construction technology'. However, it did not fare well in 'BIM applications'. Both versions were more proficient with short-answer choices than with sentence-length answers. Additionally, GPT-4 struggled with questions related to BIM policies and regulations specific to the Korean industry. Such limitations might be addressed by using tools like LangChain, which allow for feeding domain-specific documents to customize ChatGPT's responses. These advancements are anticipated to enhance ChatGPT's utility as a virtual assistant for BIM education and modeling automation.

• Journal of Gifted/Talented Education
• /
• v.20 no.3
• /
• pp.721-741
• /
• 2010

This study was to analyze verbal interactions between teachers and students after observations on teachers' questioning and feedback, students' response types and frequency analysis at middle-school class of average and gifted students. As for the verbal interaction between teachers and students of science class of general students, it was dominant for teachers to utilize questions for summarizing or guiding for textbook contents as they are. They were focused on immediate feedback in a restatement form. The students used simple responses like yes/no in general. The most high frequency of verbal interaction models expressed were in the order of cognitive-memory thinking question-short answer-immediate feedback. On the other hand, teachers of gifted students' science class threw divergent and evaluative thinking questions of open question, such as 'what's the reason?' or 'why is it?' Immediate feedback in explanatory form was mainly provided as well. The level of feedback delay was higher than general class and that of immediate feedback was lower than general class. The students preferred short words or a not-complicated sentence when they replied and their participation was more attentive and positive. Hence, The high frequency of verbal interaction models expressed were in order of cognitive-memory thinking questions－elaborative short answer－delayed feedback.

• Journal of Preventive Medicine and Public Health
• /
• v.15 no.1
• /
• pp.57-73
• /
• 1982

Questionaires on symptoms of respiratory disease have been used in Korea to elicit the probable health effects of air pollution in epidemiologic studies: The objectives of such studies often include comparing prevalence of symptoms of respiratory system between different population groups or between the same population groups, at different times. Unfortunately, little attention has, been paid to standardization of those questionaires, whether those are Korean. versions or not. Furthermore, no attempt to develop Korean ,questionaire on respiratory symptoms and relevant information has been made. Followed by 'a comparative study on responses to Korean version questionaires(English origin) of CMI, MRC, and ATS-DLD-78' two types of questionaires on respiratory symptoms and relevant information for Korean adult, which are short form (SUN-81-AS) and long forms (optional questions are added to the short one, SUN-81-AL), have been designed suitable to Korean background by authors (see Annex). The self-administered and closed-question questionaire were tested their validity and reliability by administration to l80 normal adults (medical and nursing students) and 60 clinical patients of Seoul National University Hospital, with spirometric exam. The results obtained and conclusions drawn are as follow: 1. It took less than 10 minutes to complete the questionaire SNU-81-AS and SNU-81-AL. 2. The test-retest reliability of each questions in AS and AL ,were observed as 92.7% and 92.1%, respectively. And all of the level of agreement are statistically significant with kappa statistic. 3. In addition to higher prevalence rate of symptoms in patients group compared, with, normal. group, the correlations between FEV 1.0/FVC predictive value(%) and number of symptoms were statistically significant inpatients group (See Fig. 1 and, Table 7). 4. The answer rate to optional questions in AL form among those who are not to do was about 10%, while the no-answer rate among who are to do was about 15% in Normal (medical and nursing students) group. 5. From the viewpoints of validity and reliability, the new Korean questionaire (SNU-81-AS and AL) developed by authors are to be recommendable to use in epidemiologic studies on respiratory illness in Korea. The self-administration, however, of optional questions in AL form may not assure the quality of data gathered.