• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.496-497
• /
• 2021

감정 분류는 사람의 사고방식이나 행동양식을 구분하기 위한 중요한 열쇠로, 지난 수십 년간 감정 분석과 관련된 다양한 연구가 진행되었다. 감정 분류의 품질과 정확도를 높이기 위한 방법 중 하나로 단일 레이블링 대신 다중 레이블링된 데이터 세트를 감정 분석에 활용하는 연구가 제안되었고, 본 논문에서는 T5 모델을 한국어와 영어 코퍼스로 학습한 KE-T5 모델을 기반으로 한국어 발화 데이터를 단일 레이블링한 경우와 다중 레이블링한 경우의 감정 분류 성능을 비교한 결과 다중 레이블 데이터 세트가 단일 레이블 데이터 세트보다 23.3% 더 높은 정확도를 보임을 확인했다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2020.07a
• /
• pp.673-674
• /
• 2020

의식주 중에서 자신을 표현하고 외부와의 교류를 할 수 있는 분야는 패션분야로서 인간 생활과 밀접한 관계를 가지고 있으며 사람들의 개인화된 성향 변화 및 인터넷 환경의 개선으로 트렌드는 빠르게 변화하고 있다. 인공지능 기술의 발전은 단순히 객체의 검출 및 분류에서 벗어나 패션 아이템의 분석 및 세부적인 속성을 분석할 수 있는 수준에 다다랐으며 인공지능 기술을 활용하여 사용자에게 추천할 수 있는 서비스가 출시되고 있다. 패션 트렌드의 빠른 변화 및 인공지능 기술의 발전으로 이를 활용한 플랫폼에 기반을 두어 디자이너에게는 디자인 기술을 향상시킬 수 있으며 사용자에게는 개인화된 제품을 구매할 수 있는 플랫폼 개발이 요구되고 있다. 본 논문에서는 인공지능 기술 기반 패션 분석 기술 개발을 위하여 패션 검출 모듈, 패션 검색 모듈, 패션 검색을 위한 벡터 검색 모듈, 상하의 분리를 위한 세그먼테이션 모듈, 패션 복종 분류 모듈을 개발하여 통합하였으며 패션 검색 정확도는 Top-5 기준 75.28%, 벡터 검색 속도는 벡터당 0.002m sec 이하, 세그먼테이션 추출 정확도 87.6%이상, 패션 검출 결과 IoU 0.5 환경에서 96.2%, 복종분석 90.54%의 성능을 보였다.

• Journal of Information Technology Application
• /
• v.1 no.3_4
• /
• pp.69-90
• /
• 1999

본 논문은 시계열자료의 ARMA 모형화를 위해 계층적(Hierarchical) 문제해결 방식인 인공신경망 기초 의상결정트리분류기상의 인공신경망 구조를 개선하여 지역문제(Local Problem)를 해결하는 복수개의 인공신경망 결과를 Dempster's rule of combination을 이용하여 종합하는 병행적인 (Parallel) ARMA 모형활르 위한 방법론을 제시함으로써 의사결정트리분류기에 근거한 방법론의 단점을 보완하였다. 본 논문에서 제시한 ARMA 모형화를 위한 방법론은 세 단계로 구성되어 있다: 1) ESACF 특성 벡터 추출단계; 2) 개별 인공신경망에 의한 부분적 모델링 단계; 3) Conflict Resolution 단계, 제시한 방법론을 검증하기 위해 모의실험용 자료와 실제 시계열자료를 이용하여 제시된 방법론을 검증하였으며 실험결과 기존 연구에 비해 ARMA 모형화와 정확도가 높은 것으로 나타났다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.01a
• /
• pp.345-348
• /
• 2022

본 논문에서는 유체 시뮬레이션(Fluid simulation)중 화염에서 표현되는 불똥 입자(Fire-flake particle)의 생성, 움직임과 삭제를 효율적으로 학습하고 표현할 수 있는 인공지능 기법에 대해 소개한다. 유체 시뮬레이션을 계산하기 위해서는 일반적으로 수치해석학과 같은 학문의 이해가 필요하며 불똥이나 거품과 같은 유체의 2차 효과(Secondary effect)는 기반유체(Underlying fluids)를 통해 추출되기 때문에 복잡하고 계산양이 많아진다. 이러한 문제를 완화하고자 본 논문에서는 인공신경망을 이용한 분류 모델 학습을 통해 격자 내에서 표현되어야 하는 불똥 입자의 생성을 학습하고, 다항 회귀 모델 학습을 통해 불똥 입자의 움직임을 예측한다. 또한, 불똥 입자가 삭제되어야하는 상태를 네트워크 학습을 통해 얻어내며, 수명(Lifespan) 임계값 조절하여 다양한 장면에서 불똥을 제어할 수 있다. 결과적으로 화염의 움직임을 기반으로 불똥의 움직임을 복잡한 수학식이나 디자이너에게 의존하지 않고 인공지능 학습을 통해 쉽게 제어하고 예측하는 결과를 보여준다.

• Journal of the Institute of Convergence Signal Processing
• /
• v.20 no.2
• /
• pp.78-83
• /
• 2019

Sounds based machine fault diagnosis recovers all the studies that aim to detect automatically abnormal sound on machines using the acoustic emission by these machines. Conventional methods that use mathematical models have been found inaccurate because of the complexity of the industry machinery systems and the obvious existence of nonlinear factors such as noises. Therefore, any fault diagnosis issue can be treated as a pattern recognition problem. We propose here an automatic fault diagnosis method of hand drills using discrete wavelet transform(DWT) and pattern recognition techniques such as artificial neural networks(ANN). We first conduct a filtering analysis based on DWT. The power spectral density(PSD) is performed on the wavelet subband except for the highest and lowest low frequency subband. The PSD of the wavelet coefficients are extracted as our features for classifier based on ANN the pattern recognition part. The results show that the proposed method can be effectively used not only to detect defects but also to various automatic diagnosis system based on sound.

• Journal of the Korean Society of Radiology
• /
• v.16 no.3
• /
• pp.295-302
• /
• 2022

Sparse-view computed tomography (CT) imaging technique is able to reduce radiation dose, ensure the uniformity of image characteristics among projections and suppress noise. However, the reconstructed images obtained by the sparse-view CT imaging technique suffer from severe artifacts, resulting in the distortion of image quality and internal structures. In this study, we proposed a convolutional neural network (CNN) with wavelet transformation and residual learning for reducing artifacts in sparse-view CT image, and the performance of the trained model was quantitatively analyzed. The CNN consisted of wavelet transformation, convolutional and inverse wavelet transformation layers, and input and output images were configured as sparse-view CT images and residual images, respectively. For training the CNN, the loss function was calculated by using mean squared error (MSE), and the Adam function was used as an optimizer. Result images were obtained by subtracting the residual images, which were predicted by the trained model, from sparse-view CT images. The quantitative accuracy of the result images were measured in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). The results showed that the trained model is able to improve the spatial resolution of the result images as well as reduce artifacts in sparse-view CT images effectively. Also, the trained model increased the PSNR and SSIM by 8.18% and 19.71% in comparison to the imaging model trained without wavelet transformation and residual learning, respectively. Therefore, the imaging model proposed in this study can restore the image quality of sparse-view CT image by reducing artifacts, improving spatial resolution and quantitative accuracy.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2014.04a
• /
• pp.796-797
• /
• 2014

작곡은 작곡가의 경험을 바탕으로 표현 하고자 하는 감정을 멜로디로 나타내는 창작활동이다. 본 논문에서는 이러한 작곡을 인공신경망을 이용하여 구현하기 위한 여러 가지 방법에 대하여 제안한다. 또한 제안한 방법으로 작곡된 결과를 살펴보고 결과를 분석한다. 다만 현재는 연구초기로서 추가적으로 발생하는 여러 문제를 추후 보완할 필요가 있다.

• Korea journal of population studies
• /
• v.24 no.2
• /
• pp.79-115
• /
• 2001

This research explores determinants of induced abortion in Korea. Data from the 1997 Korea National Fertility Survey by Korea Institute of Health and Social Affairs are used for this research. Another aim of this study is to verify a methodological appropriateness of survival analysis in the study of determinants of induced abortion. I compare the result of survival analysis with that of logit analysis for this purpose. In this study, it is revealed that age at conception, number of children, and number of son that have been considered as determinants of induced abortion are still significant in determining induced abortion. However, unlike results of prior research, the practice of contraception does not affect the choice of induced abortion when I control for that the conception was wanted or not. I also find that there is a significant interaction between number of son at the time of conception and intention of the conception. Induced abortion has a far reaching impact on a society as well as on the health and well-being of a woman. In Korean society, sex-selective induced abortion with son preference can cause a very high level of imbalance of sex ratio at birth. Hence, it is so important to be more cautious to the practice and prevalence of induced abortion in the age of lowe fertility.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.29 no.2B
• /
• pp.111-120
• /
• 2009

A local scour around a bridge pier is known as one of important factors of bridge collapse. Two approaches are usually used in estimating a scour depth in practice. One is to use empirical formulas, and the other is to use computational methods. But the use of empirical formulas is limited to predict a scour depth under similar conditions to which the formulas were derived. Computational methods are currently too expensive to be applied to practical engineering problems. This study presented the application of artificial neural networks (ANN) to the prediction of a scour depth around a bridge pier at an equilibrium state. This study also investigated various ANN algorithms for estimating a scour depth, such as Backpropagation Network, Radial Basis Function Network, and Generalized Regression Network. Preliminary study showed that ANN models resulted in very wide range of errors in predicting a scour depth. To solve this problem this study incorporated cluster analysis into ANN. The incorporation of cluster analysis provided better estimations of scour depth up to 42% compared with other approaches.

• Journal of Practical Engineering Education
• /
• v.15 no.1
• /
• pp.23-29
• /
• 2023

For given the importance of elementary school students developing the ability to solve problems using artificial intelligence (AI), problem-solving abilities should be developed using AI along with education to develop problem-solving abilities. Such students need a form that allows them to understand the concepts and principles of AI and to be easily educated in a fun way to understand basic understanding of how AI works. To this end, this study planned an 8-hour AI convergence program and operated based on self-driving cars, demonstrating that it was effective in improving elementary school students' problem-solving abilities, creativity, and AI understanding. As a result of operating the camp, students' understanding of AI was 3.56 (standard deviation 0.85), 4.00 (standard deviation 0.71), and t-value was -5.412 (p<0.001), indicating statistically improved understanding of AI, and high satisfaction and interest of students. In the future, it will be necessary to develop an educational program that allows elementary school students to devise their own ideas and create products to which AI models can be applied.