• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.379-384
• /
• 2022

관계 추출은 두 개체 간의 관계를 식별하는 작업이며, 비정형 텍스트를 구조화시키는 역할을 하는 작업 중 하나이다. 현재 관계 추출에서 다양한 모델에 대한 연구들이 진행되고 있지만, 한국어 관계 추출 모델에 대한 연구는 영어에 비해 부족하다. 따라서 본 논문에서는 NE(Named Entity)태그 정보가 반영된 TEM(Typed Entity Marker)과 의존 구문 그래프를 이용한 한국어 관계 추출 모델을 제안한다. 모델의 학습과 평가 말뭉치는 KLUE에서 제공하는 관계 추출 학습 말뭉치를 사용하였다. 실험 결과 제안 모델이 68.57%의 F1 점수로 실험 모델 중 가장 높은 성능을 보여 NE태그와 구문 정보가 관계 추출 성능을 향상시킬 수 있음을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2024.05a
• /
• pp.82-85
• /
• 2024

코딩 규칙 준수는 대규모 프로젝트에서 프로그램의 버그를 줄이기 위해, 또 효과적인 유지보수를 위해 필수적이나 코딩 규칙을 학습하기 위한 초보자용 도구는 거의 없는 실정이다. 본 논문에서는 Python 프로그래밍 수업에서 코딩 규칙을 학습할 수 있도록 도와주는 시스템을 제안한다. 제안된 시스템은 학습자를 위해 별도의 설치 없이 Python 코딩 규칙 검사 결과를 영어와 한글을 병행하여 출력하는 규칙 검사 뷰어를 통해 학습자의 편의성을 제공한다. 또한, 품질 점수를 계산하여 학습자의 코딩 규칙 학습의 동기를 부여한다. 제안 시스템의 성능을 평가하기 위해 SonarQube와 검출 기능을 비교하였다. 2023년도 1학기 Python 프로그래밍 수업의 제출 코드를 검사한 결과, 제안 시스템이 SonarQube보다 247% 더 많은 종류의 규칙을, 또 235% 더 많은 개수의 규칙을 검사하는 것으로 나타났다. 이러한 비교 연구 결과를 고려할 때, 제안 시스템은 학습자에게 더 나은 코딩 규칙 학습 기회를 제공할 수 있을 것으로 기대된다.

• Journal of the Korea Convergence Society
• /
• v.11 no.6
• /
• pp.1-6
• /
• 2020

In this study, research on character recognition, which is one of the fields of computer vision, was conducted. Optical character recognition, which is one of the most widely used character recognition techniques, suffers from decreasing recognition rate if the recognition target deviates from a certain standard and format. Hence, this study aimed to address this limitation by applying deep learning techniques to character recognition. In addition, as most character recognition studies have been limited to English or number recognition, the recognition range has been expanded through additional data training on Korean text. As a result, this study derived a deep learning-based character recognition algorithm for Korean text recognition. The algorithm obtained a score of 0.841 on the 1-NED evaluation method, which is a similar result to that of English recognition. Further, based on the analysis of the results, major issues with Korean text recognition and possible future study tasks are introduced.

• The Journal of the Korea Contents Association
• /
• v.14 no.5
• /
• pp.487-497
• /
• 2014

The purpose of this study is to investigate the efficacy of interaction using Wiki in communication-oriented classes taught by native teachers on the incidental improvement of productive vocabulary knowledge. Eighty two university freshmen participated in the study and they were divided into two groups. Thirty eight students(Process Group) received process feedback from the teacher while forty four students(Result Group) received result feedback. They were given a short movie project working in a group of four or five people to practice interaction between peers and between students and the teacher. Data were collected from one receptive and three productive vocabulary knowledge test scores. The results showed that interaction using Wiki was conducive to the growth of productive vocabulary knowledge incidentally due to continuous opportunities to use vocabulary. The students of process feedback group made greater gains of productive vocabulary than those of result feedback group. Based on these results, pedagogical implications are discussed.

• The Journal of the Korea Contents Association
• /
• v.14 no.10
• /
• pp.829-839
• /
• 2014

This study aims at investigating Korean EFL students' writing strategies through their rhetorical patterns and meaning-making for a writing task in an English writing classroom at a Korean university. The participants were the students and teacher in the course, and the data comprised nine pieces of students' opinion writing and interviews with the teacher. To analyze the data, a 'Claim-Support' pattern was adopted. The findings show that most students, 89%, demonstrated the same or similar elements in the 'Claim-Support' pattern for their textual structures and many parts of the meaning-making in their writing were originated from the textbook. These findings reflect that the students pursued the strategy of 'accommodation' in order to succeed in their academic writing regardless of the teacher's intention which focused on creativity and imagination in writing. The conclusion suggest that the students tend to establish their own ways of strategy to cope with the recontextualized setting for writing in English.

• Korean Journal of Cognitive Science
• /
• v.19 no.4
• /
• pp.369-396
• /
• 2008

The purpose of this article is to investigate the effect of visual working memory, the types and contents of graphic annotations on English learning. The participants of the experiments were 5th and 6th graders. The result showed that animation was effective only in the word test for children with large visual working memory, whereas text-only-annotation yielded better performance in the comprehension test in Experiment 1. The effect of relevance of animations was tested in Experiment 2. Context-relevant-animations yielded better comprehension than the animations denoting the typical meaning, whereas the opposite pattern was reported in the word test. The result of the two experiments was interpreted in terms of cognitive load theory and the generative theory of multimedia learning.

• KDI Journal of Economic Policy
• /
• v.34 no.2
• /
• pp.139-171
• /
• 2012

This paper examines the effect of private tutoring expenditures on academic performance of middle school students in South Korea, using data from "Analysis of the Level of School Education and Its Actual condition: Middle School". In the face of endogeneity of private tutoring expenditures, the paper employs an instrumental variable (IV) method and a nonparametric bounding method. Using both methods we show that the true effect of private tutoring on middle school students remains at most modest in Korea. The IV results suggest that a 10 percent increase in tutoring expenditure for Korean, English and math raises a student's test score of the subject at the largest by 1.24, 1.28, and 0.75 percent, respectively. The bounding results also fail to show evidence that an increase in tutoring expenditure leads to economically and statistically significant improvements in test score.

• The KIPS Transactions:PartB
• /
• v.14B no.3 s.113
• /
• pp.223-230
• /
• 2007

The purpose of developing an automated scoring system for English composition is to score the tests for writing English sentences and to give feedback on them without human's efforts. This paper presents an automated system to score English composition, whose input is a single sentence, not an essay. Dealing with a single sentence as an input has some advantages on comparing the input with the given answers by human teachers and giving detailed feedback to the test takers. The system has been developed and tested with the real test data collected through English tests given to the third grade students in junior high school. Two steps of the process are required to score a single sentence. The first process is analyzing the input sentence in order to detect possible errors, such as spelling errors, syntactic errors and so on. The second process is comparing the input sentence with the given answer to identify the differences as errors. The results produced by the system were then compared with those provided by human raters.

• Convergence Security Journal
• /
• v.19 no.3
• /
• pp.93-99
• /
• 2019

In the Republic From the national interest in the recent hiring scandals in the public institutions such as the Financial Supervisory Service, Gangwon Land, and so on, people were outraged by the fact that the competition opportunities were not secured although they should be equally secured to compete with anyone with their capabilities in the fair process. Hence, the government made the public institutions blind hiring to be mandatory, which eliminated writing the academic background, GPA, and English score in the application form and selected the new employees only with job competencies in the public institutions in 2017. In 2018, the prosecution did their best for the investigation on hiring scandal with the strong will to cut the scandal upon announcement of 5-year anti-corruption master plan. Amid this controversy, this study tries to suggest the further direction in the hiring military human resources upon introduction of blind hiring system, reviewing the needs to introduce blind hiring system in the military, the governmental guideline according to the mandatory blind hiring in the public institutions, and the application status of modified blind hiring in the military.

• Journal of Engineering Education Research
• /
• v.13 no.4
• /
• pp.87-92
• /
• 2010

It is strongly desirable to continuously study the program outcomes that are one of the requirements of engineering accreditation. In this paper, the results of program outcome evaluation obtained by using four tools - exit written exams, essay tests, capstone design, and an official English test - were compared to those of program outcome evaluation obtained by a course-embedded assessment method. The former evaluation was carried out for 15 students who graduated in February 2010 by using the four evaluation tools, while the latter was performed by using a course offered in the Spring and Fall semesters in 2009. Two evaluation results revealed no large difference in almost all 12 program outcomes, which indicates that course-embedded assessment would be one of the practical and effective program outcome evaluation tools.