• Journal of Korea Multimedia Society
• /
• v.21 no.12
• /
• pp.1387-1395
• /
• 2018

Core algorithm of deep learning Convolutional Neural Network(CNN) shows better performance than other machine learning algorithms. However, if there is not sufficient data, CNN can not achieve satisfactory performance even if the classifier is excellent. In this situation, it has been proven that the use of transfer learning can have a great effect. In this paper, we apply two transition learning methods(freezing, retraining) to three CNN models(ResNet-50, Inception-V3, DenseNet-121) and compare and analyze how the classification performance of CNN changes according to the methods. As a result of statistical significance test using various evaluation indicators, ResNet-50, Inception-V3, and DenseNet-121 differed by 1.18 times, 1.09 times, and 1.17 times, respectively. Based on this, we concluded that the retraining method may be more effective than the freezing method in case of transition learning in image classification problem.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.1
• /
• pp.30-45
• /
• 2024

This study introduces an effective method for predicting individual local tax delinquencies using prevalent machine learning and deep learning algorithms. The evaluation of credit risk holds great significance in the financial realm, impacting both companies and individuals. While credit risk prediction has been explored using statistical and machine learning techniques, their application to tax arrears prediction remains underexplored. We forecast individual local tax defaults in Republic of Korea using machine and deep learning algorithms, including convolutional neural networks (CNN), long short-term memory (LSTM), and sequence-to-sequence (seq2seq). Our model incorporates diverse credit and public information like loan history, delinquency records, credit card usage, and public taxation data, offering richer insights than prior studies. The results highlight the superior predictive accuracy of the CNN model. Anticipating local tax arrears more effectively could lead to efficient allocation of administrative resources. By leveraging advanced machine learning, this research offers a promising avenue for refining tax collection strategies and resource management.

• The Journal of the Korea Contents Association
• /
• v.22 no.2
• /
• pp.545-558
• /
• 2022

The purpose of this study is to explore a prediction model for accurately predicting Financial Difficulties of Chinese Cultural Industry Enterprises through Traditional Statistics and Machine Learning. To construct the prediction model, the data of 128 listed Cultural Industry Enterprises in China are used. On the basis of data groups composed of 25 explanatory variables, prediction models using Traditional Statistical such as Discriminant Analysis and logistic as well as Machine Learning such as SVM, Decision Tree and Random Forest were constructed, and Python software was used to evaluate the performance of each model. The results show that the Random Forest model has the best prediction performance, with an accuracy of 95%. The SVM model was followed with 93% accuracy. The Decision Tree model was followed with 92% accuracy.The Discriminant Analysis model was followed with 89% accuracy. The model with the lowest prediction effect was the Logistic model with an accuracy of 88%. This shows that Machine Learning model can achieve better prediction effect than Traditional Statistical model when predicting financial distress of Chinese cultural industry enterprises.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.16 no.2
• /
• pp.69-81
• /
• 2020

The objective of this study is to seek ways to maximize learning effects from e-learning by drawing improvement directions through investigating and analyzing an awareness of e-learning among e-learning attendees. The study was conducted among the attendees who are taking the e-learning program operated by K University and collected data from the students taking second semester in 2018 with the use of structured questionnaires. For data processing, SPSS Statistics 22.0 and AMOS were used, along with such analytical methods as frequency anslysis, descriptive statistical analysis, ANOVA (Analysis of Variance), t-analysis and cross tabulation. For significant data, it conducted an analysis by carrying out the Scheffe's test. According to the findings from this study, they showed a significant difference only in gender and curriculum desired to be opened in the question about e-learning participation motives per background factor. As for the learners' motives to study, it was confirmed that they tend to become more biased on time utilization and convenience of learning methods. The analysis of which factor of the three - learning factors, system factors and instructor's factors - has greatest effects on learning satisfaction indicated that learning factors influenced learning satisfaction the most in accordance with values for non-standard coefficient beta, followed by instructor factors which had a direct effect.

• The Journal of Information Systems
• /
• v.31 no.1
• /
• pp.219-238
• /
• 2022

Purpose The purpose of this study is to investigate the structural relationships between interest, interaction level, presence, which are the characteristics of metaverse, learning immersion, and learning satisfaction, which are learning factors. Design/methodology/approach A questionnaire survey technique was used to achieve the purpose of the study. A questionnaire survey was conducted from November 22 to December 5, 2021, with students with experience in non-face-to-face classes using Gather Town and a total of 114 copies of the questionnaire excluding those with insincere answers were used for empirical analysis. SPSS Win ver.23.0 was used for basic statistical analysis, and AMOS 22.0 was used for the establishment and analysis of a structural equation model. Findings According to the study findings, interest and interaction levels had effects on learning immersion and learning presence, self-efficacy on learning presence, and learning immersion and learning presence on learning satisfaction. This study is meaningful in that it conducted an empirical study to find variables for improving learning immersion by conducting classes based on metaverse. Based on the findings of this study, it was found that interest and interaction, which are the biggest characteristics of metaverse, sustain learning participation and immersion and increase presence thereby enhancing learning satisfaction so that the possibilities of metaverse as a next generation education platform passing the limit of existing real time video platforms can be peeped.

• 대한공업교육학회지
• /
• v.30 no.1
• /
• pp.56-67
• /
• 2005

The purpose of this study was to examine the effects of analytic rubrics on learners' self-directed learning ability in information technology unit assessment. The experiment and control groups were 4 classes of the 2nd grade in B middle school located in Daejeon. The pretest was performed to check the group homogeneity. For the experiment design, the nonequivalent control group design as a type of quasi-experimental design was used. The experiment was composed of 5 sessions. Statistical significance was p < .05 to verify the hypothesis, and SPSS 12.0 for Windows was used for statistical treatment. The results from this study were as follows: (1) The application of analytic rubrics in information technology unit assessment affected learners' self-directed learning ability in a positive way in 4 factor's(openness, self-concept, intrinsic motivation, self-evaluation) but does not affected in 3 factor's(autonomy, creativity, problem solving). (2) The difference in sex was not a statistically significant factor in the application of analytic rubrics in information technology unit assessment. Based on the results of the experiment, two suggestions were made to promote the application of rubrics in technology education assessment. (1) Research and development of many types of rubrics for technology education are needed. (2) Systematic training of rubrics for technology teachers is needed.

• Journal of Korean Physical Therapy Science
• /
• v.29 no.4
• /
• pp.54-64
• /
• 2022

Background: This study aims to investigate the perception and actual situation of online classes of college physical therapy students after COVID-19. In addition, it is necessary to conduct a fact-finding survey on how physical therapy students who have experienced online classes think about the online class method, what are the overall problems, and to what extent they are satisfied. Methods: The subjects of this study were 102 students in the 3rd year of physical therapy at University A, and the survey period was from June 10 to June 23, 2020, and the perception and use of online classes and self-regulated learning ability were investigated. Design: Cross-section study. Results: The perception of online lectures, it was found that the perception of online classes after actual online classes was better than the perceptions before watching (utilization), and satisfaction with online classes was generally high. There was a significant difference according to the grades in self regulated learning. The data values measured in this study were analyzed using SPSS (Statistical Package for Social Science) Windows version 12.0 statistical program. Conclusion: These results suggest that in future research, it is necessary to study the perceptions and actual conditions of each class compared to online classes and face-to-face classes.

• Journal of the Korean Society of Industry Convergence
• /
• v.26 no.4_2
• /
• pp.657-668
• /
• 2023

It's important to note that air compressors in the industrial sector are major energy consumers, accounting for a significant portion of total energy costs in manufacturing plants, ranging from 12% to 40%. To address this issue, researchers have compared forecasting models that can predict the power consumption of air compressors. The forecasting models were designed to incorporate variables such as flow rate, pressure, temperature, humidity, and dew point, utilizing statistical methods, machine learning, and deep learning techniques. The model performance was compared using measures such as RMSE, MAE and SMAPE. Out of the 21 models tested, the Elastic Net, a statistical method, proved to be the most effective in power comsumption forecasting.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.65 no.6
• /
• pp.15-22
• /
• 2023

In this study, We conduct a comparative study of deep learning-based classification of agricultural field attributes using Tagged Image File (TIF) and Enhanced Compression Wavelet (ECW) images. The goal is to interpret and classify the attributes of agricultural fields by analyzing the differences between these two image formats. "FarmMap," initiated by the Ministry of Agriculture, Food and Rural Affairs in 2014, serves as the first digital map of agricultural land in South Korea. It comprises attributes such as paddy, field, orchard, agricultural facility and ginseng cultivation areas. For the purpose of comparing deep learning-based agricultural attribute classification, we consider the location and class information of objects, as well as the attribute information of FarmMap. We utilize the ResNet-50 instance segmentation model, which is suitable for this task, to conduct simulated experiments. The comparison of agricultural attribute classification between the two images is measured in terms of accuracy. The experimental results indicate that the accuracy of TIF images is 90.44%, while that of ECW images is 91.72%. The ECW image model demonstrates approximately 1.28% higher accuracy. However, statistical validation, specifically Wilcoxon rank-sum tests, did not reveal a significant difference in accuracy between the two images.

• Communications for Statistical Applications and Methods
• /
• v.30 no.5
• /
• pp.485-499
• /
• 2023

This study aimed to predict the number of meals served in a group cafeteria using machine learning methodology. Features of the menu were created through the Word2Vec methodology and clustering, and a stacking ensemble model was constructed using Random Forest, Gradient Boosting, and CatBoost as sub-models. Results showed that CatBoost had the best performance with the ensemble model showing an 8% improvement in performance. The study also found that the date variable had the greatest influence on the number of diners in a cafeteria, followed by menu characteristics and other variables. The implications of the study include the potential for machine learning methodology to improve predictive performance and reduce food waste, as well as the removal of subjective elements in menu classification. Limitations of the research include limited data cases and a weak model structure when new menus or foreign words are not included in the learning data. Future studies should aim to address these limitations.