• Korean Journal of Child Studies
• /
• v.34 no.1
• /
• pp.123-140
• /
• 2013

The purpose of this study was to examine the effects of counting ability on young children's mathematical ability and mathematical learning potential. The subjects in this study were 75 young children of 4 & 5 years old who attended kindergartens and child care center in the city of B. They were evaluated in terms of counting ability, mathematical ability and mathematical learning potential(training and transfer) and the correlation between sub-factors and their relative influence on the partipants' mathematical ability was then analyzed. The findings of the study were as follows : First, there was a close correlation between the sub-factors of counting and those of mathematical ability. As a result of checking the relative influence of the sub-factors of counting on mathematical ability, reverse counting was revealed to have the largest impact on total mathematical ability scores and each sub-factors including algebra, number and calculation, geometry and measurement. Second, the results revealed a strong correlation between counting ability and mathematical learning ability. Regarding the size of the relative influence of the sub-factors of counting ability on training scores, reverse counting was found to be most influential, followed by continuous counting. While in relation to transfer scores, reverse counting was found to exert the greatest influence.

• Imaging Science in Dentistry
• /
• v.53 no.1
• /
• pp.27-34
• /
• 2023

Purpose: The aim of this study was to clarify the influence of training with a different kind of lesion on the performance of a target model. Materials and Methods: A total of 310 patients(211 men, 99 women; average age, 47.9±16.1 years) were selected and their panoramic images were used in this study. We created a source model using panoramic radiographs including mandibular radiolucent cyst-like lesions (radicular cyst, dentigerous cyst, odontogenic keratocyst, and ameloblastoma). The model was simulatively transferred and trained on images of Stafne's bone cavity. A learning model was created using a customized DetectNet built in the Digits version 5.0 (NVIDIA, Santa Clara, CA). Two machines(Machines A and B) with identical specifications were used to simulate transfer learning. A source model was created from the data consisting of ameloblastoma, odontogenic keratocyst, dentigerous cyst, and radicular cyst in Machine A. Thereafter, it was transferred to Machine B and trained on additional data of Stafne's bone cavity to create target models. To investigate the effect of the number of cases, we created several target models with different numbers of Stafne's bone cavity cases. Results: When the Stafne's bone cavity data were added to the training, both the detection and classification performances for this pathology improved. Even for lesions other than Stafne's bone cavity, the detection sensitivities tended to increase with the increase in the number of Stafne's bone cavities. Conclusion: This study showed that using different lesions for transfer learning improves the performance of the model.

• Journal of Digital Convergence
• /
• v.17 no.2
• /
• pp.9-18
• /
• 2019

This study investigates the factors affecting the educational satisfaction and the transfer of learning of cyber crime prevention education students in order to confirm the effectiveness of the current education. Based on the CIPP model, we confirmed whether the level of social demand and the level of knowledge in the context evaluation, recency of subjects in the input evaluation and interaction in the process evaluation affect the educational satisfaction and the transfer of learning of the students by conducting the survey for the students. As a result of analysis, it was proved that the level of knowledge, recency of subjects and interaction had a significant relationship with the educational satisfaction and recency of subjects, interaction and educational satisfaction significantly affect transfer of learning. Based on the findings, this study provides a few constructive suggestions to improve the effectiveness of the cyber crime prevention education program.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.15-30
• /
• 2023

Recently, generative models based on the Transformer architecture, such as ChatGPT, have been gaining significant attention. The Transformer architecture has been applied to various neural network models, including Google's BERT(Bidirectional Encoder Representations from Transformers) sentence generation model. In this paper, a method is proposed to create a text binary classification model for determining whether a comment on Korean movie review is positive or negative. To accomplish this, a pre-trained multilingual BERT sentence generation model is fine-tuned and transfer learned using a new Korean training dataset. To achieve this, a pre-trained BERT-Base model for multilingual sentence generation with 104 languages, 12 layers, 768 hidden, 12 attention heads, and 110M parameters is used. To change the pre-trained BERT-Base model into a text classification model, the input and output layers were fine-tuned, resulting in the creation of a new model with 178 million parameters. Using the fine-tuned model, with a maximum word count of 128, a batch size of 16, and 5 epochs, transfer learning is conducted with 10,000 training data and 5,000 testing data. A text sentiment binary classification model for Korean movie review with an accuracy of 0.9582, a loss of 0.1177, and an F1 score of 0.81 has been created. As a result of performing transfer learning with a dataset five times larger, a model with an accuracy of 0.9562, a loss of 0.1202, and an F1 score of 0.86 has been generated.

• Journal of The Korean Astronomical Society
• /
• v.52 no.6
• /
• pp.217-225
• /
• 2019

We apply a modified Convolutional Neural Network (CNN) model in conjunction with transfer learning to predict whether an active region (AR) would produce a ≥C-class or ≥M-class flare within the next 24 hours. We collect line-of-sight magnetogram samples of ARs provided by the SHARP from May 2010 to September 2018, which is a new data product from the HMI onboard the SDO. Based on these AR samples, we adopt the approach of shuffle-and-split cross-validation (CV) to build a database that includes 10 separate data sets. Each of the 10 data sets is segregated by NOAA AR number into a training and a testing data set. After training, validating, and testing our model, we compare the results with previous studies using predictive performance metrics, with a focus on the true skill statistic (TSS). The main results from this study are summarized as follows. First, to the best of our knowledge, this is the first time that the CNN model with transfer learning is used in solar physics to make binary class predictions for both ≥C-class and ≥M-class flares, without manually engineered features extracted from the observational data. Second, our model achieves relatively high scores of TSS = 0.640±0.075 and TSS = 0.526±0.052 for ≥M-class prediction and ≥C-class prediction, respectively, which is comparable to that of previous models. Third, our model also obtains quite good scores in five other metrics for both ≥C-class and ≥M-class flare prediction. Our results demonstrate that our modified CNN model with transfer learning is an effective method for flare forecasting with reasonable prediction performance.

• Culinary science and hospitality research
• /
• v.18 no.1
• /
• pp.228-245
• /
• 2012

This study attempts to investigate the relationships among psychological traits, environment, attitude on education, satisfaction with education, and learning transfer of restaurant entrepreneurship education. The samples of this study were selected from the restaurant entrepreneurs who were running restaurants after having taken the restaurant entrepreneurship education in Seoul and Kyonggi Province. Three hundred and eighty nine copies of the questionnaire, with a 86.4% response rate from a judgmental sample of 450 restaurant entrepreneurs, were utilized to study the relationships between research constructs. SPSS (11.5 version) and AMOS 5.0 were employed to analyze the uni-dimensionality of research concepts and reliability tests, and structural equation modeling was employed to verify the research hypotheses. Need for achievement and ambiguity tolerance, and environment showed a positive effect on attitude to education. Attitude to education was related positively with satisfaction with education, and satisfaction with education showed a positive effect on learning transfer of the restaurant entrepreneurship education. The managerial implications of these results were also examined.

• Journal of Industrial Convergence
• /
• v.20 no.8
• /
• pp.41-52
• /
• 2022

Education for an organization is implemented to improve the organizational and each individual's performance. However, the actual results are not as expected. Accordingly, this study is committed to investigate the education related factors that have impact on the organizational performance, which is defined by the trainee's organizational commitment and work performance. Based on the acquired knowledge, we suggest things to consider when designing corporate training for performance creation. First, it is investigated whether the task value and job relevance(educational content characteristics) and the degree of support for education within the company(organizational characteristics) affect learning-transfer of trainees. After that, the causal relationship from the learning-transfer to organizational commitment and work performance(organizational effectiveness) is analyzed. In this overall process, the effect of on-/off-line education is analyzed and compared. As a result, it is found that the task value, the job relevance, and organizational compensation have a significant impact on the learning-transfer, and the learning-transfer has impact on organization commitment and work performance. In addition, the moderating effect of the on-/off- education is identified. This study is conducted only with franchise enterprises and as a future study, a more general sampling is required to extend this work.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.16 no.4
• /
• pp.189-196
• /
• 2016

The purpose of this study is to investigate the causal relationships among self-efficacy, perceived usefulness, supervisor and peer support, satisfaction, and transfer intentions in the corporate mobile learning. For this study, the web survey was administered to 302 mobile learning learners of the A domestic corporation in South Korea. Structural equation modeling(SEM) analysis was conducted in order to examine the causal relationships among the variables. The results indicated that first, self-efficacy, perceived usefulness, and supervisor and peer support had positive effects on satisfaction. Second, supervisor and peer support and satisfaction had positive effects on transfer intentions. Third, satisfaction mediated the relationship between self-efficacy and perceived usefulness, while it did partially the relationship between supervisor and peer support and transfer intentions. Based on the result of the research, the study proposes organizational environment with cooperative supervisor and peer support should be made in order to improve the level of learners' transfer intentions. In addition, learning strategies that facilitate learners' self-efficacy and mobile information technology acceptance are needed to develop for enhancing the learners' satisfaction.

• Journal of Broadcast Engineering
• /
• v.26 no.6
• /
• pp.809-812
• /
• 2021

Air bubbles which may be generated during the PCB coating process can be a major cause of malfunction. so it is necessary to detect the bubbles in advance. In previous studies, candidates for bubbles were extracted using the brightness characteristics of bubbles, and the candidates were verified using CNN(Convolutional Neural Networks). In this paper, we propose a bubble detection method using a transfer learning-based CNN model. The VGGNet is adopted and sigmoid is used as a classification layer, and the last convolutional layer and classification layer are trained together when transfer learning is applied. The performance of the proposed method is F1-score 0.9044, which shows an improvement of about 0.17 compared to the previous study.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.3
• /
• pp.433-441
• /
• 2021

Android applications are released across various categories, including productivity apps and games, and users are exposed to various applications and even malware depending on their usage patterns. On the other hand, most analysis engines train using existing datasets and do not reflect user patterns even if periodic updates are made. Thus, the detection rate for known malware is high, while types of malware such as adware are difficult to detect. In addition, existing engines incur increased service provider costs due to the cost of server farm, and the user layer suffers from problems where availability and real-timeness are not guaranteed. To address these problems, we propose an analysis system that performs on-device malware detection through transfer learning, which requires only one-time communication with the server. In addition, The system has a complete process on the device, including decompiler, which can distribute the load of the server system. As an evaluation result, it shows 90.3% accuracy without transfer learning, while the model transferred with adware catergories shows 95.1% of accuracy, which is 4.8% higher compare to original model.