• Asia pacific journal of information systems
• /
• v.32 no.2
• /
• pp.226-248
• /
• 2022

The existing model for sentiment analysis of product reviews learned from past data and new data was labeled based on training. But new data was never used by the existing system for making a decision. The proposed Aspect-based multi-agent Deep Reinforcement learning Sentiment Analysis (ADRSA) model learned from its very first data without the help of any training dataset and labeled a sentence with aspect category and sentiment polarity. It keeps on learning from the new data and updates its knowledge for improving its intelligence. The decision of the proposed system changed over time based on the new data. So, the accuracy of the sentiment analysis using deep reinforcement learning was improved over supervised learning and unsupervised learning methods. Hence, the sentiments of premium customers on a particular site can be explored to other customers effectively. A dynamic environment with a strong knowledge base can help the system to remember the sentences and usage State Action Reward State Action (SARSA) algorithm with Bidirectional Encoder Representations from Transformers (BERT) model improved the performance of the proposed system in terms of accuracy when compared to the state of art methods.

• Geomechanics and Engineering
• /
• v.38 no.1
• /
• pp.29-44
• /
• 2024

This research investigates the effectiveness of data augmentation techniques in the automated analysis of B-scan images from ground-penetrating radar (GPR) using deep learning. In spite of the growing interest in automating GPR data analysis and advancements in deep learning for image classification and object detection, many deep learning-based GPR data analysis studies have been limited by the availability of large, diverse GPR datasets. Data augmentation techniques are widely used in deep learning to improve model performance. In this study, we applied four data augmentation techniques (geometric transformation, color-space transformation, noise injection, and applying kernel filter) to the GPR datasets obtained from a testbed. A deep learning model for GPR data analysis was developed using three models (Faster R-CNN ResNet, SSD ResNet, and EfficientDet) based on transfer learning. It was found that data augmentation significantly enhances model performance across all cases, with the mAP and AR for the Faster R-CNN ResNet model increasing by approximately 4%, achieving a maximum mAP (Intersection over Union = 0.5:1.0) of 87.5% and maximum AR of 90.5%. These results highlight the importance of data augmentation in improving the robustness and accuracy of deep learning models for GPR B-scan analysis. The enhanced detection capabilities achieved through these techniques contribute to more reliable subsurface investigations in geotechnical engineering.

• Journal of Fisheries and Marine Sciences Education
• /
• v.21 no.1
• /
• pp.16-27
• /
• 2009

The purpose of this study was to develop of learning strategy scale for college students. This study further classified several sub-areas and defined each concepts of learning strategy. Based upon the classification of each sub-areas, tentative test items were developed through the verification of validity by three professionals. A pilot study of the developed scale was administered to 239 college students. And the research collected major data from 1,012 college students. Data were analyzed to obtain item quality, reliability, and validity analysis. The results of this study were as follows. The scale for learning strategy was defined by eight factors and they were 'self-management strategy', 'examination-readiness strategy', 'cognitive strategy', 'memorizing strategy', 'reporting strategy', 'resource-utilization strategy', 'self-regulated strategy', 'cooperative learning strategy'. The results of the confirmatory factor analysis proved the eight factors in the learning strategy. And criterion validity evidence was also obtained from a correlation analysis of the level of academic achievement.

• Journal of Fisheries and Marine Sciences Education
• /
• v.26 no.1
• /
• pp.148-155
• /
• 2014

The purpose of this research was to analyze the effects of smart learning. By using meta analysis method, twenty MA and Ph.D degree papers published from 2006 to 2013 were analyzed and 104 effect sizes were calculated. Followings were the results of the research: (a) Smart learning turned out to be more statistically effective comparing to traditional education. The total mean effect size was .886 and the value of U3 was 66.53%. (b) All effect size of sub dependent variables(ie, academic achievement, learning satisfaction, learning attitude) were also effective by adapting smart learning. (c) The moderated variables likes learner characteristics, learning content, and interaction had high effect sizes. Operation system variable had a low effect size but it was not significant.

• International Journal of Advanced Culture Technology
• /
• v.7 no.4
• /
• pp.295-301
• /
• 2019

Since the third artificial intelligence boom was triggered by deep learning, it has been 10 years. It is time to analyze and discuss the research trends of deep learning for the stable development of AI. In this regard, this study systematically analyzes the trends of research on deep learning over the past 10 years. We collected research literature on deep learning and performed LDA based topic modeling analysis. We analyzed trends by topic over 10 years. We have also identified differences among the major research countries, China, the United States, South Korea, and United Kingdom. The results of this study will provide insights into research direction on deep learning in the future, and provide implications for the stable development strategy of deep learning.

• Journal of the Korea Society of Computer and Information
• /
• v.21 no.7
• /
• pp.93-100
• /
• 2016

This paper shows what kind of influence the learning motivation factors have on the effectiveness of Flipped Learning Model through the case of operating a JAVA programming subject. The Flipped Learning Approach consisting of Before Class, Before or At Start of Class, and In Class provides the students with learning motivation as well as satisfies Keller's ARCS(Attention, Relevance, Confidence, Satisfaction) to keep them studying steadily. This research conducts the operation of Flipped Learning and gets Exploratory Factor Analysis and Reliability Analysis from the result of the course experience questionnaire at the end of the class. Given this survey result, Flipped Learning approach improves the learners' satisfaction in class and the effectiveness in the fields of understanding learning context more than does the previous lecture-based learning approach by pacing learning procedure and conducting self-directed learning.

• Journal of Fisheries and Marine Sciences Education
• /
• v.24 no.6
• /
• pp.935-949
• /
• 2012

The purpose of this study was to investigate the structural relationship among the self-efficacy, self-directed learning ability, school adjustment and learning flow in middle school students by the structural equation modeling analysis. The subjects of this study consisted of 553 middle school students. The data were analyzed with descriptive statistics, Pearson correlations and structural equation modeling analysis by using the SPSS 12.0 and AMOS 5.0 statistical program. The results of this study were as followed: First, there were significant correlations among the self-efficacy, self-directed learning ability, school adjustment and learning flow. Second, the self-directed learning ability and school adjustment directly affected the learning flow. Third, self-efficacy and school adjustment variables indirectly affected learning flow. The indices of the best fit model on these variable were adequate. This study shows that the self-efficacy, self-directed learning ability, school adjustment are the significant predictor for the learning flow during adolescent.

• Journal of information and communication convergence engineering
• /
• v.18 no.4
• /
• pp.239-244
• /
• 2020

This study analyzed the influencing factors of learning engagement and teaching presence in online programming practice classes. The subjects of this study were students enrolled in an industrial specialized high school, who practiced creating Arduino circuits and programming using a web-based virtual practice tool called Tinkercad. This research adopted a tool that can measure task value, learning flow, learning engagement, and teaching presence. Based on this analysis, learning flow had a mediating effect between task value and online learning engagement, as well as between task value and teaching presence. Increasing learning engagement in online classes requires sensitizing the learners about task value, using hands-on platforms available online, and expanding interaction with instructors to increase learning flow of students. Furthermore, using virtual hands-on tools in online programming classes is relevant in increasing learning engagement. Future research tasks include: confirming the effectiveness of online learning engagement and teaching presence through pre- and post-tests, and conducting research on various practical subjects.

• Journal of Information Technology Services
• /
• v.21 no.3
• /
• pp.89-103
• /
• 2022

The purpose of this study is to explore professors' perception of blended learning. For this purpose, the reflective journals written by 56 university professors was analyzed using the keyword network analysis method. The results of this study are as follows: First, as a result of keyword frequency analysis for the blended learning, the keywords showed the highest frequency in the order of (1) 'instructional design', 'student', 'instructional method', 'learning objective' in the area of learning, (2) 'importance', 'instruction', 'feeling', 'student' in the area of feeling, and (3) 'semester', 'plan', 'weekly', and 'instruction' in the area of action plan. Second, the results of analyzing the degree, closeness centrality, and betweenness centrality of network connection are as follows. (1) The keywords 'instruction', 'instructional method', 'instructional design', and 'learning objective' in the area of learning, (2) the keywords 'instruction', 'importance', and 'necessity' in the area of feeling, and (3) 'instruction', 'plan', and 'semester' in the area of action plan showed high values in degree, closeness centrality, and betweenness centrality. Based on the research results, implications for blended learning and professors' perception were discussed.

• International Journal of Internet, Broadcasting and Communication
• /
• v.10 no.2
• /
• pp.60-67
• /
• 2018

The purpose of this study was to examine the effects of internet-use Anatomy class, as one of the Flipped learning method, on critical thinking disposition. The class for this study was conducted from March 1 to April 10, 2018. The study involved a total of 180 people in the first year of a University located in C province. Data collection was carried out before and after the Flipped learning method application. Frequency analysis, Paired t-test, Pearson correlation, and Regression analysis were used for the analysis. According to the analysis, 28.3% of men and 71.1% of women and before applying the program analysis of correlation between Flipped learning perception and critical thinking disposition showed a significant correlation between confidence(sub-component of critical thinking) only (p<.005). Comparing the scores of critical thinking before and after the program, it was found that Truth seeking (p<.001), Open-mindness (p<.005), Confidence (p<.001), Systematicity (p<.005), Analyticity (p<.001), and Inquisitiveness (p<.001) scores had increased significantly except Maturity (p>.005). And the regression analysis of Flipped learning method applying influence on critical thinking disposition were significantly affected (p<.001). Based on the results of this study, it was possible to determine that Flipped learning method had a positive effect on critical thinking disposition.