• Broadcasting and Media Magazine
• /
• v.27 no.2
• /
• pp.19-25
• /
• 2022

본 고에서는 최근 컴퓨터 비전 분야에서 가장 활발히 연구되고 있는 분야 중에 하나인 자가 학습(Self-supervised Learning) 기술의 동향과 향후 방향성에 대해서 논의한다. 컴퓨터 비전 분야에서의 자가 학습 기술은 최근에 Contrastive Learning 기법을 활용하여 활발하게 연구되고 있는데, 이를 위한 좋은 Positive와 Negative를 어떻게 추출할까에 대한 고민으로 수많은 연구들이 진행되어 왔다. 본 고에서는 이러한 방향성에서 대표적인 몇 가지의 방법론에 대해서 논의하고 이의 한계점을 언급하며 컴퓨터 비전 분야에서 자가 학습 기법이 가야 할 방향성에 대해서 논의하고자 한다.

• Journal of Broadcast Engineering
• /
• v.28 no.1
• /
• pp.109-123
• /
• 2023

Deep learning has made rapid progress in recent years and is affecting various fields and industries. The art field cannot be an exception, and in this paper, we would like to explore and experiment and analyze research fields that creatively generate 2D images in 3D from a visual arts and engineering perspective. To this end, the original image of the domestic artist is learned through GAN or Diffusion Models, and then converted into 3D using 3D conversion software and deep learning. And we compare the results with prior algorithms. After that, we will analyze the problems and improvements of 2D to 3D creative generation.

• Smart Media Journal
• /
• v.13 no.6
• /
• pp.35-43
• /
• 2024

During the post-pandemic period, the interest in foreign language learning surged, leading to increased usage of language-learning apps. With the rising demand for these apps, analyzing app reviews becomes essential, as they provide valuable insights into user experiences and suggestions for improvement. This research focuses on extracting insights into users' opinions, sentiments, and overall satisfaction from reviews of HelloTalk, one of the most renowned language-learning apps. We employed topic modeling and emotion analysis approaches to analyze reviews collected from the Google Play Store. Several experiments were conducted to evaluate the performance of sentiment classification models with different settings. In addition, we identified dominant emotions and topics within the app reviews using feature importance analysis. The experimental results show that the Random Forest model with topics and emotions outperforms other approaches in accuracy, recall, and F1 score. The findings reveal that topics emphasizing language learning and community interactions, as well as the use of language learning tools and the learning experience, are prominent. Moreover, the emotions of 'admiration' and 'annoyance' emerge as significant factors across all models. This research highlights that incorporating emotion scores into the model and utilizing a broader range of emotion labels enhances model performance.

• Journal of Broadcast Engineering
• /
• v.25 no.2
• /
• pp.275-278
• /
• 2020

In this paper, we proposed a deep learning based character recognition system through virtual data generation. In order to secure the learning data that takes the largest weight in supervised learning, virtual data was created. Also, after creating virtual data, data generalization was performed to cope with various data by using augmentation parameter. Finally, the learning data composition generated data by assigning various values to augmentation parameter and font parameter. Test data for measuring the character recognition performance was constructed by cropping the text area from the actual image data. The test data was augmented considering the image distortion that may occur in real environment. Deep learning algorithm uses YOLO v3 which performs detection in real time. Inference result outputs the final detection result through post-processing.

• 전자공학회논문지 IE
• /
• v.49 no.1
• /
• pp.30-38
• /
• 2012

The participants were 102 students with digital broadcasting and media major.. A hypothetical model proposed included learning flow, school support as predictors, and satisfaction and learning persistence as a criterion. The results of this study through multiple regression analysis indicated that learning flow and school flow predicted significantly on satisfaction. And school flow, school support, and satisfaction predicted significantly on learning persistence. In addition, satisfaction mediated between learning flow and learning persistence, and between school flow and learning persistence. A constructive foundation for providing learning strategies in the successful engineering education would be proposed on the basis of the current results of this study.

• Smart Media Journal
• /
• v.12 no.1
• /
• pp.24-31
• /
• 2023

Meta-learning is metacognition that instantly distinguishes between knowing and unknown. It is a learning method that adapts and solves new problems by self-learning with a small amount of data.A few-shot learning method is a type of meta-learning method that accurately predicts query data even with a very small support set. In this study, we propose a method to solve the limitations of the prototype created with the mean-point vector of each class. For this purpose, we use the few-shot learning method that created the prototype used in the few-shot learning method as the median prototype. For quantitative evaluation, a handwriting recognition dataset and mini-Imagenet dataset were used and compared with the existing method. Through the experimental results, it was confirmed that the performance was improved compared to the existing method.

• Industry Promotion Research
• /
• v.2 no.2
• /
• pp.31-44
• /
• 2017

This study is to investigate the relationship between the communication type recognized by members of organization and the learning transfer climate in the organization, and when it comes to the type of communications within organization set as independent variable, four variables such as communication with supervisor, media quality communication, and organizational outlook communication were proposed and regarding learning transfer climate which is a dependent variable, four sub-parameters such as supervisor support, colleague support, transfer opportunity and organizational compensation recognition were selected. 116 copies of the statistical data excluding statistically insignificant surveys were used for 150 participants who participated in the 2015 core competency curriculum for SMEs and 3 months have passed, in order to achieve this study's objective. Meanwhile, the research hypothesis was verified with the collected data through frequency analysis, factor analysis, reliability verification, technical statistical analysis, and simple and multiple regression analysis by using the statistical package program of SPSS 20.0. Results of this study have shown that firstly, communication with supervisor, media quality communication, and organizational outlook communication all showed a positive (+) significant influence on the superior support in the influence relationship between communication type in organization and supervisor support which is learning transfer climate. Secondly, communication type in organization doesn't appear to affect colleague support of learning transfer climate. Thirdly, communication with supervisor, media quality communication, and organizational outlook except for communication with colleagues have shown a positive (+) influence on transfer opportunity in the influence relationship between communication type in organization and transfer opportunity of learning transfer climate. Lastly, communication with supervisor and communication on organizational outlook showed positive(+) influence in the influence relationship between communication type in organization and organizational compensation recognition of learning transfer climate.

• Korean Journal of Culture and Social Issue
• /
• v.16 no.2
• /
• pp.179-195
• /
• 2010

Media reports of suicides has been found to increase suicide cases that were temporally and spacially proximal to the reports, but the psychological mechanisms, social learning, underlying the negative effects was not directly tested. The present study examined the cognitive processes of social learning that media reports of suicides, especially positive contents toward suicides, might change people's perception, memory, and attitudes toward suicides positively and subsequently increase subsequent suicide intentions and behaviors. Through an internet survey, 300 adults reported their perception, memory, and attitudes toward news reports of suicides, and rated whether the suicides were described positively or negatively in the reports. Finally they reported their suicide intentions and behaviors. The results revealed that people tended to remember more the contents of suicide reports suggested to increase copycat suicides. Also, people were found to have an ironic view to suicide reports of media that they acknowledged the dangers of suicides reports and approached the reports with curiosity. More importantly, the perception of the positive reward that suicides might achieved through suicides was related with positive attitudes toward suicides and behavioral intention to suicides. The present findings was discussed in the social learning understanding of copycat suicides and their implications for suicide-prevention strategies.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.11
• /
• pp.1456-1462
• /
• 2016

In this paper, we apply a chord progression learning model to a rule-based generation of a four-part chorus. The proposed system is given a 32-note melody line and completes the four-part chorus based on the rule of harmonics, predicting the chord progression with the CRBM model. The data for the training model was collected from various harmony textbooks, and chord progressions were extracted with key-independent features so as to utilize the given data effectively. It was shown that the output piece obtained with the proposed learning model had a more natural progression than the piece that used only the rule-based approach.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.8
• /
• pp.958-964
• /
• 2016

In this paper, we propose an i-vector based language recognition system to identify the spoken language of the speaker, which uses multiple discriminative deep neural network (DNN) models analogous to the multi-class support vector machine (SVM) classification system. The proposed model was trained and tested using the i-vectors included in the NIST 2015 i-vector Machine Learning Challenge database, and shown to outperform the conventional language recognition methods such as cosine distance, SVM and softmax NN classifier in open-set experiments.