• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.606-610
• /
• 2009

We propose a method utilizing BLEU(BiLingual Evaluation Understudy), which is widely used in automatic evaluation of machine translations, for an autonomous assessment of a short essay answer. BLEU evaluates translations with an assumption that the translation by a machine is supposed to be more accurate as it is getting to be more similar to the translation by a human. BLEU scores the translation by comparing the n-grams of translations by a machine and humans. Similarly we score students answers by comparing to multiple reference answers with BLEU. In the experiment, we compute correlation coefficient values between scores of our system and human instructors.

• Journal of Educational Research in Mathematics
• /
• v.23 no.4
• /
• pp.533-551
• /
• 2013

The purpose of this study is to investigate how middle school mathematics teachers understand the definition and the purpose of constructed-response assessments as well as what they expect from the assessments. By examining the responses of 189 middle school mathematics teachers to the survey, we found that there is discrepancy between their perceptions and reality of the constructed-response assessments in practice. In other words, teachers' practice is not well aligned with their perceptions and expectations due to the restraint of school system and environment. The findings give us some meaningful implication for the improvement of effective constructed-response assessments and professional development.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.11a
• /
• pp.813-816
• /
• 2005

본 논문에서는 한문 문장의 기본구조를 자연어 처리과정을 이용하여 채점하는 시스템을 구현하고자 하였다. 기존의 주관식 채점시스템은 단답형이 대부분이며, 서술형은 문장에서 키워드를 추출하여 유사도를 비교한 후 채점하는 방식으로 키워드 추출 시 명사류만 대상으로 하여 다른 품사들은 누락되며, 또한 문장의 구조를 간과하는 문제점이 있다. 본 시스템은 자연어 처리과정을 이용한 것으로 어휘분석과 구문분석을 통해 문장구조를 확인하고, 문장성분별로 해당 단어를 정확하게 사용하였는가를 검사하여 채점함으로써 기존의 문제점을 개선한다.

• Communications of Mathematical Education
• /
• v.26 no.3
• /
• pp.273-299
• /
• 2012

The purpose of this study is to investigate how constructed-response assessments are currently defined and used to find an alternative way of developing well-defined constructed-response questions and an assessment system. Based on our review of the literature, we analyze constructed-response assessments developed and used in Gyeonggi-do, South Korea and Ohio, U.S.A. in terms of their definitions, types of questions, information and guidelines given by the government agencies in each place, and scoring rubrics and their use. The results provide meaningful implications for the development of constructed-response assessments in the future.

• Journal of the Korean School Mathematics Society
• /
• v.11 no.3
• /
• pp.377-397
• /
• 2008

In today's secondary school mathematics curriculum, assessment is to be focused on evaluating the student's mathematical thinking rather than finding the correct solution to the problem. A descriptive evaluation method is therefore introduced to the school districts and suggested as an alternative assessment method in K-12 school mathematics. Descriptive evaluation method is widely available for use in schools, but there are barriers to using it since the teachers are forced to use the method by the school districts. In this research, we surveyed 120 secondary school mathematics teachers in Seoul and the surrounding metropolitan area to understand the status and the perception about using descriptive evaluation. The goal of the study was to find and understand the direct implications of using the new assessment method in secondary mathematics classes. The study showed that most of the mathematics teachers used the descriptive assessment method during their regular exam periods which is given twice per semester. Most of the open-ended problems used for descriptive evaluation were medium or high level math questions which were graded by the teachers at least 3 times to ensure objective evaluation. The teachers stated that objectivity in grading, administrative situation, and development of descriptive problems were the difficult barriers in descriptive evaluation. The teachers in the survey also commented that teachers' administrative responsibility should be reduced and that school environment in general should be improved for the new assessment method to become successful. Finally, the study showed that development of more descriptive problems with specific grading guidelines need to be developed for each grade level.

• The Mathematical Education
• /
• v.63 no.2
• /
• pp.165-186
• /
• 2024

Despite the growing attention on artificial intelligence-based automated scoring technology as a support method for the introduction of descriptive items in school environments and large-scale assessments, there is a noticeable lack of foundational research in mathematics compared to other subjects. This study developed an automated scoring model for two descriptive items in first-year middle school mathematics using the Random Forest algorithm, evaluated its performance, and explored ways to enhance this performance. The accuracy of the final models for the two items was found to be between 0.95 to 1.00 and 0.73 to 0.89, respectively, which is relatively high compared to automated scoring models in other subjects. We discovered that the strategic selection of the number of evaluation categories, taking into account the amount of data, is crucial for the effective development and performance of automated scoring models. Additionally, text preprocessing by mathematics education experts proved effective in improving both the performance and interpretability of the automated scoring model. Selecting a vectorization method that matches the characteristics of the items and data was identified as one way to enhance model performance. Furthermore, we confirmed that oversampling is a useful method to supplement performance in situations where practical limitations hinder balanced data collection. To enhance educational utility, further research is needed on how to utilize feature importance derived from the Random Forest-based automated scoring model to generate useful information for teaching and learning, such as feedback. This study is significant as foundational research in the field of mathematics descriptive automatic scoring, and there is a need for various subsequent studies through close collaboration between AI experts and math education experts.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.1027-1030
• /
• 2005

기존 의미커널을 적용한 주관식 채점 시스템은 여러 답안과 말뭉치에서 추출한 색인어들과의 상관관계를 벡터방식으로 표현하여 자연어 처리에 대한 문제를 해결하려 하였다. 본 논문에서는 기존 시스템의 답안 및 색인어의 표현 한계로 인한 유사도 계산오차 가능성에 대한 문제를 해결하고자 시소러스를 이용한 임의 추출 방식의 답안 확장을 적용하였다. 서술형 주관식 평가에서는 문장의 문맥보다는 사용된 어휘에 채점가중치가 높다는 점을 착안, 출제자와 수험자 모두의 답안을 동의어, 유의어 그룹으로 확장하여 채점 성능을 향상시키려 하였다. 우선 두 답안을 형태소 분석기를 이용해 색인어를 추출한 후 워드넷을 이용하여 동의어, 유의어 그룹으로 확장한다. 이들을 말뭉치 색인을 이용하여 단어들 간 상관관계를 측정하기 위한 벡터로 구성하고 의미 커널을 적용하여 정답 유사도를 계산하였다. 출제자의 채점결과와 각 모델의 채점 점수의 상관계수 계산 결과 ELSA 모델이 가장 높은 유사도를 나타내었다..

• Journal of Science Education
• /
• v.46 no.1
• /
• pp.121-149
• /
• 2022

Descriptive assessment is a meaningful assessment method in relation to problem solving ability, reasoning ability, and communication ability as emphasized in mathematics curriculum. In Korea, as performance assessment has been emphasized since the 7th mathematics curriculum, descriptive assessment is being conducted as a method of performance assessment in schools. However, descriptive assessment has not been introduced in the university scholastic ability test for various reasons. Considering that descriptive assessment is emphasized in the mathematics classroom and has sufficient educational value, a serious discussion on the implementation of descriptive assessment in the university scholastic ability test will be necessary. In this study, we analyzed the descriptive assessment of Russia's unified state examination (USE) in the mathematics, which corresponds to Korea's university scholastic ability test. Through a literature review, we investigated how mathematics examination problems were structured in the USE and which mathematical abilities were required for the examination. In particular, the outer structure of the problems was analyzed focusing on the mathematics problems of the USE 2021, and the scoring method of the descriptive problems was also analyzed. The results of this study are expected to provide a variety of information on the possibility of introducing descriptive assessment in the Korean university scholastic ability tests.

• Journal of Korean Home Economics Education Association
• /
• v.23 no.3
• /
• pp.69-90
• /
• 2011

To develop the descriptive evaluation questionnaire with high validity and reliability on the Clothing and Textiles section of the middle school Technology & Hone Economics textbook, apply it to students and analyze its results. We made out a draft for descriptive evaluation questionnaire that was based upon the concrete establishment of the goal and the range of evaluation. We also made a rubric for scoring as well as sample answer-sheets. Finally, we completed a total of twenty three descriptive evaluation questions and we applied it to sixty five 2nd-grade students in two classes in a middle school. Descriptive evaluation questionnaire exhibited the relative high validity on each question. Moreover, three graders gave the same score on each question of descriptive evaluation, suggesting that descriptive evaluation questionnaire has the high inter-grader reliability and the strong correlation. But, low academic achievement was generally observed in the subjects. They had difficulty in describing their knowledge via their own language and drawing up accurate and detailed answers. They recognized the positive aspects of descriptive evaluation questionnaire, but they felt it uncomfortable due to study-burden and description itself. To overcome these limitations, it is required that students should experience various materials related to subject contents in classes as well as textbooks, concentrate themselves on finding solutions for problems, expand their scope, and practice describe them in advance. Therefore, the additional training for description evaluation questionnaire will be necessary for the more efficient and discriminative questionnaire. Also the questionnaire with high validity and reliability should be developed and the aggressive and voluntary participation of teachers will be needed.

• Journal of Science Education
• /
• v.38 no.1
• /
• pp.168-181
• /
• 2014

The purpose of this study was to monitor evaluation in science education at primary and secondary school in Korea. In order to do this, the survey was administered. The subject was 518 teachers; 292 elementary school teachers, 110 middle school science teachers and 116 general high school science teachers. It was found that the ratio of paper examination increased while that of performance evaluation decreased according to the grade of school. In the science paper examination, the ratio of selection type item was higher than that of supply type item, and the ratio of item belonging to knowledge area was higher than that of item belonging to the inquiry area regardless of the grade of school, teaching career, and teacher's gender. The teachers recognized the necessity of supply type item, but they suffered from the reliability of the result of that item. And it was hard for them to do performance evaluation because it took much time and effort. This study suggested that the high-quality in-service teacher training be provided and laboratory assistants who help the science experiment and grading test papers be employed in order to fulfil the performance evaluation and increase the ratio of supply type item and items belonging to the inquiry area.