• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.22 no.5
• /
• pp.75-86
• /
• 2022

This study aims to solve the problem of class imbalance in numerical data by using a deep learning-based Variational AutoEncoder and to improve the performance of the learning model by augmenting the learning data. We propose 'D-VAE' to artificially increase the number of records for a given table data. The main features of the proposed technique go through discretization and feature selection in the preprocessing process to optimize the data. In the discretization process, K-means are applied and grouped, and then converted into one-hot vectors by one-hot encoding technique. Subsequently, for memory efficiency, sample data are generated with Variational AutoEncoder using only features that help predict with RFECV among feature selection techniques. To verify the performance of the proposed model, we demonstrate its validity by conducting experiments by data augmentation ratio.

• Journal of Convergence for Information Technology
• /
• v.9 no.2
• /
• pp.108-114
• /
• 2019

In this paper, we propose an intelligent vocabulary recommendation agent that automatically provides vocabulary corresponding to game-based learners' needs and requirements in the mobile education augmented reality game environment. The proposed agent reflects the characteristics of mobile technology and augmented reality technology as much as possible. In addition, this agent includes a vocabulary reasoning module, a single game vocabulary recommendation module, a battle game vocabulary recommendation module, a learning vocabulary list Module, and a thesaurus module. As a result, game-based learners' are generally satisfied. The precision of context vocabulary reasoning and thesaurus is 4.01 and 4.11, respectively, which shows that vocabulary related to situation of game-based learner is extracted. However, In the case of satisfaction, battle game vocabulary(3.86) is relatively low compared to single game vocabulary(3.94) because it recommends vocabulary that can be used jointly among recommendation vocabulary of individual learners.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2021.07a
• /
• pp.137-139
• /
• 2021

본 논문에서는 목조 문화재 영상에서의 변위를 효율적으로 감지하기 위한 객체 중심 증강 기법을 사용한 모델을 제안한다. 우선 객체 중심 증강 기법을 적용하여 변위 객체들이 이미지 공간상의 어느 곳이든 위치할 수 있게끔 데이터를 구성한 이후 사전 학습된 합성 곱 신경망을 사용하여 입력 이미지에 대한 심층 특징 벡터를 추출한다. 그 이후 심층 특징 벡터는 완전 연결 계층의 입력 값으로 들어와서 최종적으로 변위가 존재하는지 아닌지에 대한 예측을 수행하게 된다. 데이터 셋으로는 충주시 근처의 문화재에 방문해서 수집한 목조 문화재 이미지를 가지고 정상 및 비정상으로 구분한 데이터 셋을 사용하였다. 실험 결과 우리가 제안한 객체 중심 증강 기법을 사용한 모델이 객체 중심 증강 기법을 사용하지 않은 모델보다 목조 문화재에서 변위 영역을 더 잘 감지함을 확인하였다. 이러한 결과로부터 우리가 제안한 방법이 목재 문화재의 변위 검출에 있어서 매우 적합함을 보여준다.

• Annual Conference of KIPS
• /
• 2023.11a
• /
• pp.1102-1103
• /
• 2023

안전모 착용 여부를 확인하는 객체 탐지 모델을 물류 현장에서 활용하기 위해서는 안전모를 착용한 경우와 착용하지 않은 경우를 정확하게 탐지해야 한다. 하지만 학습 데이터가 안전모를 착용한 클래스와 착용하지 않은 클래스 간 불균형이 존재하는 경우 해당 데이터만으로는 태스크에 맞게 학습이됐다고 보긴 힘들다. 본 연구는 데이터 증강 기법 적용 시 임의의 데이터에 증강을 적용하는 대신 상대적으로 적은 안전모를 착용하지 않은 클래스를 포함하는 이미지에 대하여 데이터 증강 기법을 적용하였다. 여러 데이터 증강 기법 중 Rotation, Gaussian Noise, 객체를 기준으로 한 Crop을 직접 구현 및 적용하여 객체 탐지 모델인 YOLOv5의 성능을 효과적으로 높이며 더욱 강건한 모델을 개발하는 방법을 제안한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.10
• /
• pp.455-460
• /
• 2023

Text augmentation is a methodology that creates new augmented texts by transforming or generating original texts for the purpose of improving the performance of NLP models. However existing text augmentation techniques have limitations such as lack of expressive diversity semantic distortion and limited number of augmented texts. Recently text augmentation using large language models and few-shot learning can overcome these limitations but there is also a risk of noise generation due to incorrect generation. In this paper, we propose a text augmentation method called TAGS that generates multiple candidate texts and selects the appropriate text as the augmented text. TAGS generates various expressions using few-shot learning while effectively selecting suitable data even with a small amount of original text by using contrastive learning and similarity comparison. We applied this method to task-oriented chatbot data and achieved more than sixty times quantitative improvement. We also analyzed the generated texts to confirm that they produced semantically and expressively diverse texts compared to the original texts. Moreover, we trained and evaluated a classification model using the augmented texts and showed that it improved the performance by more than 0.1915, confirming that it helps to improve the actual model performance.

• Proceedings of the Korean Information Science Society Conference
• /
• 2004.10a
• /
• pp.268-270
• /
• 2004

본 논문은 증강 현실(augmented reality)을 이용하여 학습자에게 효과적으로 교육할 수 있는 e-learning 교육컨텐츠 환경을 제공한다. 현재까지 연구되어온 증강환경을 이용한 교육 컨텐츠는 단순한 상호작용만을 허용하고, 실감형 인터페이스를 제공하기에는 제약이 따른다. 본 논문은 이를 개선하기 위해 기존의 교육 컨텐츠를 실감형 교육 컨텐츠로 전환할 수 있는 시스템 구성 방법에 관한 것으로써. 컨텐츠의 배치, 조작, 학습과정 등에서 기존의 교육 컨텐츠보다 인터랙티브하고 직관적으로 교육할 수 있는 실감형 인터페이스를 제공한다.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.6
• /
• pp.301-307
• /
• 2023

Deep learning requires building a deep neural network, collecting a large amount of training data, and then training the built neural network for a long time. If training does not proceed properly or overfitting occurs, training will fail. When using deep learning tools that have been developed so far, it takes a lot of time to collect training data and learn. However, due to the rapid advent of the mobile environment and the increase in sensor data, the demand for real-time deep learning technology that can dramatically reduce the time required for neural network learning is rapidly increasing. In this study, a real-time deep learning system was implemented using an Arduino system equipped with a fine dust sensor. In the implemented system, fine dust data is measured every 30 seconds, and when up to 120 are accumulated, learning is performed using the previously accumulated data and the newly accumulated data as a dataset. The neural network for learning was composed of one input layer, one hidden layer, and one output. To evaluate the performance of the implemented system, learning time and root mean square error (RMSE) were measured. As a result of the experiment, the average learning error was 0.04053796, and the average learning time of one epoch was about 3,447 seconds.

• The Journal of the Korea Contents Association
• /
• v.10 no.8
• /
• pp.461-471
• /
• 2010

Recently, Augmented Reality(AR) technologies have been emerged, which shows the types of digital contents integrating real and virtual worlds. To maximize the effect of AR technology, tangible user interface, which enables users to interact with the contents in the same way in which they manipulate objects in real world, is applied. In particular, we expect that the technologies are able to enhance learners' interests and degree of immersion, and produce new learning contents in order to maximize the effect of learning. In this paper, we propose a learning system for scientific experiments with multi-touch screen and tangible user interface. The system consists of an experiment table equipped with a large multi-touch screen and a realistic learning device that can detect the user's simple gestures. In real world, some scientific experiments involve high cost, long time or dangerous objects, but this system will overcome such hindrance and provide learners with a variety of experiment experience in realistic ways.

• Communications of Mathematical Education
• /
• v.33 no.3
• /
• pp.207-229
• /
• 2019

In the era of the Fourth Industrial Revolution, various attempts have been made to incorporate ICT technology into mathematics teaching and learning, and the necessity and efficiency of classroom instruction using flipped learning, virtual reality and augmented reality have attracted attention. This leads to an increase in demand for instructional contents and their use in education. Therefore, there is a growing need for the development of instructional contents that can be applied in the field and the study of teaching methods. In this point of view, this research classifies the types of teaching-learning, presents the flipped learning instruction and mathematics contents by teaching-learning types using constructivist mathematics education principles and augmented reality-based mobile applications. These methods and lesson plans can provide a useful framework for teaching-learning in mathematics education.

• Journal of The Korean Association For Science Education
• /
• v.40 no.2
• /
• pp.191-201
• /
• 2020

This study investigated verbal and physical interactions which appeared in collaborative science concept learning using augmented reality. Twelve 10th grade students participated in this study. After being organized into three four-member small groups, they participated in classes using smart device-based augmented reality application developed for the understanding of the chemical bonding concept. Their class activities were audio- and video-taped. Semi-structured interviews were also conducted. The results revealed that within individual statement units of verbal interaction, the proportions of information question/explanation and direction question/explanation were found to be high. Within interaction units, the proportions of reformative and cumulative interaction were relatively high. The proportions of progress were also found to be high within both individual statement units and interaction units of verbal interaction. Students' physical interactions were mainly conducted without meaningful verbal interactions. When their physical interactions were accompanied by knowledge construction-related verbal interactions, the proportions of gazing virtual objects and worksheet-related interactions were high. In contrast, various exploratory activities related to the manipulation of markers mainly appeared when they conducted physical interactions only, or when their physical interactions were accompanied by management-related verbal interactions. On the bases of the results, effective methods for collaborative concept learning using augmented reality in science education are discussed.