• Journal of Science Education
• /
• v.37 no.2
• /
• pp.348-358
• /
• 2013

The purpose of this study is to develop the science item on computer-based performance assessment which is able to replace the assessment with actual experimental apparatus. The survey on the necessities and the difficulties of the assessment with actual experimental apparatus was conducted with 57 physics teachers. By reflecting the results of the survey, the computer-based assessment items which are related with constructing circuits with an ammeter and a voltmeter were developed. The developed computer-based assessment items were organized in similarity to assessment conducted with actual experimental apparatus, proper feedback was provided by students' performance in the process of assessment so as to grading performance results automatically. To achieve them, the algorithm was developed so that computer can judge the accuracy of students' performance results.

• Proceedings of the Korean Nuclear Society Conference
• /
• 1998.05a
• /
• pp.297-302
• /
• 1998

최근 원자력산업계 동향은 기기의 노후화, 예비품 확보의 어려움, 그리고 기존 설비에 대한 보수비용 증가 등의 이유로 아날로그 계측제어 설비들을 컴퓨터-기반 설비로의 부분적인 개선이나 전체적인 교체가 추진되고 있다 그러나 컴퓨터-기반 설비는 소프트웨어와 하드웨어가 환경영향에 민감하고, 공통모드고장을 일으킬 수 있는 프로그래밍 설계오류의 잠재성이 있는 것으로 알려져 있다. 공통모드고장을 방지할 수 있는 가장 설득력 있는 해결방안은 철저한 품질보증, 심층방어 및 다양성 설계기법으로 평가되고 있다. 본 논문에서는 컴퓨터-기반 계측제어계통의 심층방어 및 다양성에 관한 규제기준과 정성적 평가를 위한 블록개념을 소개하고, 평가의 사례로서 CE System 80+와 국내 가동중인 W형 원전의 설비개선에서 주급수 상실시 컴퓨터-기반 설비의 공통 모드고장에 따른 다양성과 심층방어계층을 평가하였다. 특히 국내 가동중인 W형 원전에 대해서는 그 평가결과를 근거로 하여 ATWS 완화설비의 설치를 제안하였다.

• Journal of the Korean School Mathematics Society
• /
• v.20 no.3
• /
• pp.325-344
• /
• 2017

Since PISA2006, the computer based assessment in mathematics(CBAM) was introduced for the first times and at last PISA2015 used all items in CBAM for problem solving. In this study, we focused on which important properties were considered in constructing geometric 'fence items' used in PISA 2015 to find the future direction over our teacher education, especially for constructing 'computer based assessment items.' For the purpose of the study, we analyzed the fence items on three components such as dependency, invariant, and path found in dragging activities, within a computer environment using the dynamic Geometry Software, GSP. Also, for the future, we provided an open-ended problem related to the fence items, which we could use as the merit of computer-based environment.

• Journal of The Korean Association For Science Education
• /
• v.43 no.1
• /
• pp.17-28
• /
• 2023

In line with the entry into the digital-based intelligent information society, the science curriculum emphasizes the cultivation of scientific competencies, and computer-based test (CBT) is drawing attention for assessment of competencies. CBT has advantages to develop items that have high fidelity, and to establish a feedback system by accumulating results into the database. However, it is necessary to solve the problems of improving validity of assessment results, lowering measurement efficiency, and increasing management factors. To examine students' responses to the introduction of the new assessment tools in the process of transitioning from paper-based test (PBT) to CBT, in this study, we analyzed the results of the PBT and the CBT conducted in 2021 National Assessment of Educational Achievement (NAEA). In particular, we sought to find the effects on student achievement when only the mode of assessment was changed without change of items, and the effect on student achievement when the items were composed including technology enhanced features that take advantage of CBT. This study is derived through the analysis of the results of 7,137 third-grade middle school students taking one among the three kinds of assessments, which were the PBT or two kinds of CBT. After the assessment, the percentage of correct answers and the item discriminations were collected for each group, and expert opinions on characteristics of response were collected through the expert council involving 8 science teachers with experience in NAEA. According to the results, there was no significant difference between students' achievement results in the PBT and the CBT-M, which means simple mode conversion type of CBT, so it could be explained that the mode effect did not appear. However, it was confirmed that the percentage of correct answers for the construct response items was somewhat high in the CBT, and this result was analyzed to be related to the convenience of the response. On the other hand, there were the items with a difference of more than 10%p from the correct answer rate of similar items, among the items to which technology enhanced functions were applied following the introduction of CBT. According to the analysis of response rate of options, these results could be explained that the students' level of understanding could be more closely grasped through the innovative items developed through the technology enhanced function. Based on the results, we discussed some guidance to be considered when introducing CBT and developing items through CBT, and presented implications.

• Journal of Internet Computing and Services
• /
• v.17 no.1
• /
• pp.73-81
• /
• 2016

This paper deals with the practical experience and lessons to develop a method of course-embedded assessment for program outcomes in Computer Science & Information Engineering. In engineering education, ABEEK(Accreditation Board for Engineering Education of Korea) has been nominated as an official accreditation from Korean government under some conditions of fulfillment. One of the conditions is enforcement of course-embedded assessment for program outcomes. In this paper, a method of course-embedded assessment in Computer Science & Information Engineering is suggested. The case study was done in a basic project course which includes the special assessment methods in project courses. The case study of course-embedded assessment in a basic project course was given with consideration for each step, that is, set up of program outcomes which are highly related to the course, selection of assessment tools such as written examination and project-based assessment, set up of assessment criteria, implementation of course-embedded assessment, and analysis of results. The proposed method is expected to be a good example to construct the whole strategy for course-embedded assessment in Computer Science & Information Engineering.

• Proceedings of The KACE
• /
• 2017.08a
• /
• pp.93-97
• /
• 2017

오늘날의 교육은 단순히 지식을 많이 습득하는 것보다 문제를 인식하고 지식 및 정보를 통합하여 창의적으로 문제를 해결할 수 있는 역량 중심의 교육을 강화하고 있다. 이러한 역량을 배양하고 함양하기 위해 학생들의 발달 수준을 알아보고 분석하기 위해 국제 수준의 학업 성취도 평가 연구가 국제기구를 중심으로 활발히 이루어지고 있다. 학습자들의 역량을 효과적으로 평가하기 위해 최근에 컴퓨터 기반 평가가 도입되고 있다. 본 연구에서는 국제 수준의 학업 성취도 평가에서 컴퓨터 기반 평가가 이루어지는 현황을 파악하고 PISA와 ICILS의 공개 문항의 특징을 비교하여 시사점을 제시하였다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.176-181
• /
• 2022

일반 상식 추론 능력은 가장 사람다운 능력 중 하나로써, 인공지능 모델이 쉽게 모사하기 어려운 영역이다. 딥러닝 기반의 언어 모델은 여전히 일반 상식에 기반한 추론을 필요로 하는 분야에서 부족한 성능을 보인다. 특히, 한국어에서는 일반 상식 추론과 관련한 연구가 상당히 부족한 상황이다. 이러한 문제 완화를 위해 최근 생성 기반의 일반 상식 추론을 위한 한국어 데이터셋인 Korean CommonGen [1]이 발표되었다. 그러나, 해당 데이터셋의 평가 지표는 어휘 단계의 유사성과 중첩에 의존하는 한계를 지니며, 생성한 문장이 일반 상식에 부합한 문장인지 측정하기 어렵다. 따라서 본 논문은 한국어 일반 상식 추론 및 생성 능력에 대한 평가 지표를 개선하기 위해 문장 성분의 의미역과 자모의 형태 변화를 바탕으로 생성 결과를 평가하는 SRLev, 사람의 평가 결과를 학습한 BIH, 그리고 두 평가 지표의 장점을 결합한 SRLev-BIH를 제안한다.

• Convergence Security Journal
• /
• v.18 no.2
• /
• pp.33-40
• /
• 2018

ICT technology has been applied to various educational fields, but applying to educational test field is limited. Computer-based test (CBT) can overcome temporal and spatial constraints of conventional paper-based test, but is vulnerable to fraud by test parties. In this paper, we propose real-time monitoring and process management methods to enhance the security of CBT. In the proposed methods, the test screens of students are periodically captured and transferred to the professor screen to enable real-time monitoring, and the possible processes used for cheating can be blocked before testing. In order to monitor the screen of many students in real time, effective compression of the captured original image is important. We applied three-step compression methods: initial image compression, resolution reduction, and re-compression. Through this, the original image of about 6MB was converted into the storage image of about 3.8KB. We use the process extraction and management functions of Windows API to block the processes that may be used for cheating. The CBT system of this paper with the new security enhancement methods shows the superiority through comparison of the security related functions with the existing CBT systems.

• Journal of Educational Research in Mathematics
• /
• v.27 no.2
• /
• pp.157-170
• /
• 2017

Mathematics was the main domain of PISA 2012, and both paper-based and computer-based assessment of mathematics (CBAM) were conducted. PISA 2012 was the first large-scale computer-based mathematics assessment in Korea, and it is meaningful in that it evaluated students' mathematical literacy in problem situations using dynamic geometry, graph, and spreadsheet. Although Korea ranked third in CBAM, the use of ICT in mathematics lessons appeared to be low. On the other hand, this study focused on Singapore, which ranked first in CBAM. The Singapore Ministry of Education developed online programs such as AlgeTools and AlgeDisc, and implemented the programs in classes by specifying them in mathematics curriculum and textbooks. Thus, this study investigated influential factors on computer-based assessment of mathematics by comparing the results of Korea and Singapore, and aimed to provide meaningful evidence on the direction of Korea's ICT-based mathematics education. The results showed that ICT use at home for school related tasks, attitudes towards computers as a tool for school learning, and openness and perseverance of problem solving were positively associated with computer-based mathematics performance, whereas the use of ICT in mathematics class by teacher demonstration was negatively related. Efforts are needed to improve computer use and enhance teaching techniques related to ICT use in Korean math classes. Future research is recommended to examine how effectively teachers use ICT in mathematics class in Singapore.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.9-13
• /
• 2023

본 논문에서는 대규모 언어모델의 검색 기반 답변 생성능력을 평가하는 새로운 한국어 벤치마크, KFREB(Korean Fictional Retrieval Evaluation Benchmark)를 제안한다. KFREB는 모델이 사전학습 되지 않은 허구의 정보를 바탕으로 검색 기반 답변 생성 능력을 평가함으로써, 기존의 대규모 언어모델이 사전학습에서 보았던 사실을 반영하여 생성하는 답변이 실제 검색 기반 답변 시스템에서의 능력을 제대로 평가할 수 없다는 문제를 해결하고자 한다. 제안된 KFREB는 검색기반 대규모 언어모델의 실제 서비스 케이스를 고려하여 장문 문서, 두 개의 정답을 포함한 골드 문서, 한 개의 골드 문서와 유사 방해 문서 키워드 유무, 그리고 문서 간 상호 참조를 요구하는 상호참조 멀티홉 리즈닝 경우 등에 대한 평가 케이스를 제공하며, 이를 통해 대규모 언어모델의 적절한 선택과 실제 서비스 활용에 대한 인사이트를 제공할 수 있을 것이다.