• ETRI Journal
• /
• v.46 no.1
• /
• pp.59-70
• /
• 2024

We introduce a high-performance named entity recognition (NER) model for written and spoken language. To overcome challenges related to labeled data scarcity and domain shifts, we use transfer learning to leverage our previously developed KorBERT as the base model. We also adopt a meta-pseudo-label method using a teacher/student framework with labeled and unlabeled data. Our model presents two modifications. First, the student model is updated with an average loss from both human- and pseudo-labeled data. Second, the influence of noisy pseudo-labeled data is mitigated by considering feedback scores and updating the teacher model only when below a threshold (0.0005). We achieve the target NER performance in the spoken language domain and improve that in the written language domain by proposing a straightforward rollback method that reverts to the best model based on scarce human-labeled data. Further improvement is achieved by adjusting the label vector weights in the named entity dictionary.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.6
• /
• pp.81-89
• /
• 2018

In this paper, we propose to investigate the existing EPL education related studies and to collect the total effect size for the improvement of high-level thinking through meta-analysis and to confirm the effect size according to various variables. So, we have objectively identified and generalized the practical effects of EPL education on the various elements of high-level thinking and high-level thinking. The results of the meta-analysis showed that 1) EPL is a teaching-learning tool that greatly improves students' high-level thinking. 2) Education based EPL has greatly contributed to the enhancement of creative thinking and logical thinking among high-level thinking. 3) Kodu, App Inventor, Scratch, and Dolittle was confirmed that the effect on the improvement of high-level thinking was great.

• Journal of the Korean Society for Library and Information Science
• /
• v.52 no.2
• /
• pp.191-220
• /
• 2018

Using Bloom's taxonomy model of thinking ability for learning (RBTT) and new ACRL information literacy framework (2016), this study demonstrates the meta-literacy competencies for library specialists and analyses current LIS curriculum of higher education. Some guidelines for reformation of LIS curriculum, emphasizing meta-literacy competencies which are required from Web3.0 information environment, are provided.

• Journal of the Korea Society of Computer and Information
• /
• v.23 no.11
• /
• pp.239-246
• /
• 2018

In this paper, we propose to investigate the existing SW education related studies and to collect the total effect size for the improvement of computational thinking(CT) through meta-analysis and to confirm the effect size according to various variables. So, we have objectively identified and generalized the practical effects of SW education on the various variable. The results of the meta-analysis showed that 1) the overall effect of SW educational activities on CT improvement was 'Hedges' g=0.643', so SW educational activity can get a CT improvement of about 24% over other educational activities. 2) EPL is a SW teaching-learning method that greatly improves students' CT. 3) Since SW education can be guaranteed to improve CT among elementary school students and university students, essential SW education for these should be further promoted. And it was found that SW education activity were more effective for CT improvement in small cities & rural areas than big cities, but the cause was not identified, so further study will be required.

• Research in Mathematical Education
• /
• v.24 no.3
• /
• pp.175-199
• /
• 2021

The Science Writing Heuristic (SWH) approach is described as an immersive argument-based science inquiry focusing particularly on learning through epistemic practices. In the literature, several previous studies indicate how academic achievement is positively influenced by the SWH. In addition to these previous studies, several meta-syntheses of qualitative data have been conducted on this particular topic. With these literatures in mind, a quantitative meta-analysis was conducted with ten studies (N = 724) to examine the effectiveness of the SWH on student achievement in Turkey. To present a thoroughly detailed report, this study also examined the following moderators: grade level, subject area, school location, intervention length, and report source. Overall, this study found that in Turkey, the SWH classrooms performed better in academic achievement tests than traditional lecture-based classrooms. Additionally, the SWH is more likely to be effective regardless of grade levels, subject areas, and school locations.

• Journal of information and communication convergence engineering
• /
• v.22 no.2
• /
• pp.153-158
• /
• 2024

This study aims to summarize the effects of Arduino-based educational programs through a meta-analysis. Eleven eligible primary studies were obtained through a systematic literature review and coded accordingly. The results are as follows: The meta-analysis revealed that the overall effect size for all the studies was 0.518. Analysis of the moderator variables indicated statistically significant differences between them. Regarding the learning domains, the results were ranked in descending order of the cognitive and affective domains. Within the cognitive domain, the effect sizes were ranked in descending order as follows: logical thinking, content knowledge, convergence competency, self-efficacy, computational thinking, and creative problem-solving skills. In terms of subject areas, the descending order of effect sizes was agriculture, STEAM, environmental science, practical arts, artificial intelligence, informatics, and computers. Regarding school level, the results were ranked in the following descending order: college, elementary school, middle school, and high school.

• Journal of Industrial Convergence
• /
• v.21 no.10
• /
• pp.111-120
• /
• 2023

This study is a descriptive research to examine the effects of meta-cognition and resilience on clinical reasoning ability of nursing students who have completed the simulation education integrated with problem based learning. The study subjects were senior nursing students who had experienced SIM-PBL education, and data was collected by using a structured questionnaire from September to December 2021. The collected data was analyzed employing descriptive statistics, correlation, and hierarchical regression analysis using the SPSS program. The results demonstrated that meta-cognition and resilience had a significant positive correlation with clinical reasoning ability. The chief factors influencing on the clinical reasoning ability of nursing students were as follows: confidence in participating in the SIM-PBL education, meta-cognition, and resilience. In addition, the three factors explained the clinical reasoning ability at a high level of 75%. The clinical reasoning ability of nursing students may be cultivated by applying internal reinforcers of self-confidence, meta-cognition, and resilience into a SIM-PBL simulation.

• Journal of Educational Research in Mathematics
• /
• v.25 no.1
• /
• pp.113-137
• /
• 2015

The objective of the present study is to systematically examine the effects of on the cognitive and affective domains of elementary, middle, and high school students by conducting a meta-analysis. To this end, this study selected 31 research papers that had analyzed the effects of applying, and performed a meta-analysis of the findings presented in each research paper. The results obtained from the meta-analysis are presented as follows. First, both the collaborative learning program and the peer tutoring program for underachieving students in math manifested an above average size of effect in the cognitive domain. In particular, the effect was the greatest at the elementary school level, and out of the two programs, peer tutoring was identified to have a sizable effect. Second, both programs displayed an above average size of effect in the affective domain, and peer tutoring was identified to have a higher effect than collaborative learning. In addition, when the programs were compared based on school levels, the size of effect was highest at the elementary school level followed by middle school and high school, in that order. When compared based on the criteria of the affective domain, self-efficacy in math, learners' attitude toward math, and learners' interest in math were identified to. Finally, this study presented suggestions for teaching underachieving students in math and conducting follow-up studies based on the analysis results.

• Journal of The Korean Association For Science Education
• /
• v.38 no.6
• /
• pp.875-883
• /
• 2018

This study examined overall effect of STEAM programs on student learning through meta-analysis of journal articles published for the past six years. We examined the areas of effects that the research tested and analyzed overall effect across the research. We first identified academic journal articles that utilized quasi-experimental design in examining STEAM effects on student learning and presented appropriate data for meta-analysis such as effect size. A total of 63 articles were identified to be appropriate for meta-analysis. Using R packages, we first identified outliers and eliminated them in the analysis of mean effect size. Thus, 172 effect sizes from 60 studies were analyzed. The results showed that the mean effect was medium (effect size = 0.52). Analysis showed that moderators of the effect were affective measures, thinking skills, character measures, and career aspirations, which meant the studies that measured these variables had more effect than achievement measures. On the other hand, the school level (elementary, middle, and high school), the absence or presence of student products as program requirements, hours of intervention, and sample size did not moderate the effect. Thus, regardless of these variables STEAM programs produced medium effect in general. Based on these results, further research areas and topics are suggested.

• The Journal of Korean Association of Computer Education
• /
• v.14 no.3
• /
• pp.25-36
• /
• 2011

The purpose of this study was to analyze the effects of educational programming language(EPL) using the meta-analysis method. In order to achieve the purpose of this study, t-test and F-test were performed for the effect size differences between the variables. The results of the study were as follows: First, EPL turned out to be highly effective in improving learning effects. The total mean of effect size was as big as 1.01 and the value of $U_3$ was 84.38%. EPL increased the learning effect by 34.38% compared with the control group. Second, the moderator variables such as subject, publication type, and learner's school age there was no statistically significant differences. By designing the experiment nonequivalent control group pretest-posttest design showed statistically significant effect size compared with single group pretest-posttest design. Third, the mean effect sizes of the dependent variables were as follows: Creativity 1.90, problem solving ability 1.25, logical thinking ability 1.18, learning motivation 0.81, and achievement 0.59. EPL showed positive effect than traditional teaching and learning method comprehensively.