• 대한간호학회지
• /
• 제29권3호
• /
• pp.700-710
• /
• 1999

Promoting women's health lifestyles are important due to their connection to family health. The purpose of this study was to analyse women's health lifestyles(HPL) and their effects on women's life-cycle, so in order to develop a program in a women's health care center. The subjects included were 1080 women over 18 years old living in Wonju city, and were selected by stratified and purposive sampling. The data were collected through a questionnaire and interview. The Cronbach $\alpha$, %, mean, ANOVA, Pearson's correlation, and regression in SPSS PC Win. package was used to analyze the data. The sample was sepernted into three groups premarital group 20.2%(premarital women between 18 and 40 years old), delivery and children rearing group 49.9%(marital women between 18 and 40 years old), over middle agedelderly group 29.9%(women over 41 years old). Significant difference were found in the HPL according to group. Also, relating variables, such as self-efficacy, family functions, health locus of control and gender role perception that were considered relating variables to HPL significantly differed among the three groups. HPL significantly correlated with self-efficacy, family functions, HLOC and gender role perception in all participants and at all groups. The regression analysis of HPL was interpreted 40.6% with relating variables, self-efficacy, health attention, family functions, and internal locus of control, health perception, power other locus of control and chance locus of control in all participant. Self-efficacy, family functions, health attention were considered important variables in premarital group, self-efficacy, family functions, internal locus of control, health attention, health perception and power of control were important in delivery-rearing group. Self-efficacy, health attention, internal locus of control, family functions and health perception were important in middle aged-elderly group. As a result, we found the differences HPL scores and relating variables according to life-cycle groups. Therefore, we should prepare health promoting education programs for women according to women's life cycles. Also we suggest that women's health care centers based on communities was needed for proper management of women's health.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• ETRI Journal
• /
• 제44권3호
• /
• pp.438-449
• /
• 2022

This study proposes a dependency grammar-based self-attention multilayered bidirectional long short-term memory (DG-M-Bi-LSTM) model for subject-predicate-object (SPO) tuple recognition from natural language (NL) sentences. To add recent knowledge to the knowledge base autonomously, it is essential to extract knowledge from numerous NL data. Therefore, this study proposes a high-accuracy SPO tuple recognition model that requires a small amount of learning data to extract knowledge from NL sentences. The accuracy of SPO tuple recognition using DG-M-Bi-LSTM is compared with that using NL-based self-attention multilayered bidirectional LSTM, DG-based bidirectional encoder representations from transformers (BERT), and NL-based BERT to evaluate its effectiveness. The DG-M-Bi-LSTM model achieves the best results in terms of recognition accuracy for extracting SPO tuples from NL sentences even if it has fewer deep neural network (DNN) parameters than BERT. In particular, its accuracy is better than that of BERT when the learning data are limited. Additionally, its pretrained DNN parameters can be applied to other domains because it learns the structural relations in NL sentences.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2021년도 춘계학술발표대회
• /
• pp.343-346
• /
• 2021

약물 디자인이란 단백질과 같은 생물학적 표적에 작용할 수 있는 새로운 약물을 개발하는 과정이다. 전통적인 방법은 탐색과 개발 단계로 구성되어 있으나, 하나의 신약 개발을 위해서는 10 년 이상의 장시간이 요구되기 때문에, 이러한 기간을 단축하기 위한 인공지능 기반의 약물 디자인 방법들이 개발되고 있다. 하지만 많은 심층학습 기반의 약물 디자인 모델들은 RNN 기법을 활용하고 있고, RNN 은 훈련속도가 느리다는 단점이 있기 때문에 개선의 여지가 남아있다. 이런 단점을 극복하기 위해 본 연구는 self-attention 과 variational autoencoder 를 활용한 SMILES 생성 모델을 제안한다. 제안된 모델은 최신 약물 디자인 모델 대비 훈련 시간을 1/36 단축하고, 뿐만 아니라 유효한 SMILES 를 더 많이 생성하는 것을 확인하였다.

• 한국산학기술학회논문지
• /
• 제16권1호
• /
• pp.429-437
• /
• 2015

본 연구는 대학생의 스마트폰 중독 경향에 따른 신체활동량, 수면, 주의력 조절 및 자기조절학습과의 차이와 관계를 파악하기 위한 서술적 조사연구이다. 자료수집은 269명의 대학생을 대상으로 구조화된 설문지를 통해 조사하였고 SPSS 18.0 프로그램을 통해 자료를 분석하였다. 연구결과 대상자의 스마트폰 중독수준은 성별에 따라 차이가 있었고, 스마트폰 중독성향이 강할수록 성적과 자가통제력은 낮고, 스마트폰 사용 시간이 길었다. 스마트폰 중독수준과 신체활동량, 수면의 질 및 주의력 조절능력이 유의한 차이가 있었고, 스마트폰 중독정도가 높을수록 신체활동량과 자기조절학습능력 및 수면의 질이 낮은 경향이 있었고, 주의력 조절은 높게 나타나는 경향을 보였다. 이상의 결과를 통해 대학생의 스마트폰의 과다사용으로 일상적 건강과 학습능력이 저하될 수 있으며 이를 방지하기 위한 다양한 차원에서의 스마트폰 중독예방 전략이 필요함을 확인하였다.

• 한국음향학회지
• /
• 제39권5호
• /
• pp.483-489
• /
• 2020

본 논문에서는 저자원 환경의 음성인식에서 음향 모델의 성능을 높이기 위한 음향 모델 학습 방법을 제안한다. 저자원 환경이란, 음향 모델에서 100시간 미만의 학습 자료를 사용한 환경을 말한다. 저자원 환경의 음성인식에서는 음향 모델이 유사한 발음들을 잘 구분하지 못하는 문제가 발생한다. 예를 들면, 파열음 /d/와 /t/, 파열음 /g/와 /k/, 파찰음 /z/와 /ch/ 등의 발음은 저자원 환경에서 잘 구분하지 못한다. 자기 주의 메커니즘은 깊은 신경망 모델로부터 출력된 벡터에 대해 가중치를 부여하며, 이를 통해 저자원 환경에서 발생할 수 있는 유사한 발음 오류 문제를 해결한다. 음향 모델에서 좋은 성능을 보이는 Time Delay Neural Network(TDNN)과 Output gate Projected Gated Recurrent Unit(OPGRU)의 혼합 모델에 자기 주의 기반 학습 방법을 적용했을 때, 51.6 h 분량의 학습 자료를 사용한 한국어 음향 모델에 대하여 단어 오류율 기준 5.98 %의 성능을 보여 기존 기술 대비 0.74 %의 절대적 성능 개선을 보였다.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.1-16
• /
• 2022

Despite recent breakthroughs in deep learning and computer vision fields, the pixel-wise identification of tiny objects in high-resolution images with complex disturbances remains challenging. This study proposes a modified U-net for tiny crack segmentation in real-world steel-box-girder bridges. The modified U-net adopts the common U-net framework and a novel Self-Attention-Self-Adaption (SASA) neuron as the fundamental computing element. The Self-Attention module applies softmax and gate operations to obtain the attention vector. It enables the neuron to focus on the most significant receptive fields when processing large-scale feature maps. The Self-Adaption module consists of a multiplayer perceptron subnet and achieves deeper feature extraction inside a single neuron. For data augmentation, a grid-based crack random elastic deformation (CRED) algorithm is designed to enrich the diversities and irregular shapes of distributed cracks. Grid-based uniform control nodes are first set on both input images and binary labels, random offsets are then employed on these control nodes, and bilinear interpolation is performed for the rest pixels. The proposed SASA neuron and CRED algorithm are simultaneously deployed to train the modified U-net. 200 raw images with a high resolution of 4928 × 3264 are collected, 160 for training and the rest 40 for the test. 512 × 512 patches are generated from the original images by a sliding window with an overlap of 256 as inputs. Results show that the average IoU between the recognized and ground-truth cracks reaches 0.409, which is 29.8% higher than the regular U-net. A five-fold cross-validation study is performed to verify that the proposed method is robust to different training and test images. Ablation experiments further demonstrate the effectiveness of the proposed SASA neuron and CRED algorithm. Promotions of the average IoU individually utilizing the SASA and CRED module add up to the final promotion of the full model, indicating that the SASA and CRED modules contribute to the different stages of model and data in the training process.

• 한국가정과교육학회지
• /
• 제28권2호
• /
• pp.95-111
• /
• 2016

본 연구의 목적은 청소년기 학교폭력 경험이 자아정체감에 미치는 영향을 알아보고, 학교폭력 경험과 자아정체감의 관계에서 가족 내 사회자본에 따른 조절효과가 있는 지 분석하는 데 있다. 분석을 위해 한국아동 청소년패널(KCYPS) 중에서 중1패널 3차년도(2012) 데이터를 활용하였다. 분석에 사용한 통계프로그램은 IBM SPSS Statistics 22.0 version이었으며, 기술통계, 피어슨 상관 계수, 독립 t검정, 위계적 회귀분석을 실시하였다. 본 연구의 연구결과는 다음과 같다. 첫째, 자아정체감과 학교폭력 경험(비행가해 및 피해), 가족 내 사회자본(부모의 관심애정 및 친구인지)의 평균을 분석한 결과 자아정체감, 가족 내 사회자본 하위요인인 부모의 관심애정과 친구인지는 평균보다 다소 높았고, 학교폭력 경험 하위요인인 비행 가해경험과 피해경험은 적은 것으로 조사되었다. 둘째, 자아정체감, 학교폭력 경험(비행가해 및 피해), 가족 내 사회자본(부모의 관심애정 및 친구인지)의 상관관계를 분석한 결과 비행피해, 부모의 관심애정, 부모의 친구인지가 자아정체감에 유의한 상관관계가 있는 것으로 나타났다. 셋째, 회귀 분석 결과 통제변수 중에서는 성별, 모친의 최종학력, 또래애착이 자아정체감의 영향력을 높이며 유의미하였고, 학교폭력 경험 하위요인인 비행가해는 정적 관계로, 비행피해는 부적 관계로 유의미한 영향력을 가졌으며, 가족 내 사회자본의 하위요인인 부모의 관심애정과 친구인지도 모두 정적 관계로 자아정체감의 영향력을 높이며 통계적으로 유의하였다. 넷째, 학교폭력 경험(비행가해 및 피해)과 자아정체감의 관계에서 가족 내 사회자본(부모의 관심애정 및 친구인지)의 조절효과에 대한 분석결과 비행가해에 대한 부모의 관심애정, 비행가해에 대한 부모의 친구인지, 비행피해에 대한 부모의 친구인지의 상호작용항을 투입했을 때는 조절효과가 나타나지 않았고, 비행피해와 부모의 관심애정 상호작용 항을 투입했을 때는 부(-)의 상호작용 효과로 나타났다.

• Journal of Information Processing Systems
• /
• 제17권3호
• /
• pp.453-461
• /
• 2021

Many deep learning approaches have been studied for image classification in computer vision. However, there are not enough data to generate accurate models in medical fields, and many datasets are not annotated. This study presents a new method that can use both unlabeled and labeled data. The proposed method is applied to classify cervix images into normal versus cancerous, and we demonstrate the results. First, we use a patch self-supervised learning for training the global context of the image using an unlabeled image dataset. Second, we generate a classifier model by using the transferred knowledge from self-supervised learning. We also apply attention learning to capture the local features of the image. The combined method provides better performance than state-of-the-art approaches in accuracy and sensitivity.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제7권11호
• /
• pp.2824-2838
• /
• 2013

There has been a lot of attention paid recently to analyze dynamic human gestures that vary over time. Most attention to dynamic gestures concerns with spatio-temporal features, as compared to analyzing each frame of gestures separately. For accurate dynamic gesture recognition, motion feature extraction algorithms need to find representative features that uniquely identify time-varying gestures. This paper proposes a new feature-extraction algorithm using temporal self-similarity based on a hierarchical human model. Because a conventional temporal self-similarity method computes a whole movement among the continuous frames, the conventional temporal self-similarity method cannot recognize different gestures with the same amount of movement. The proposed model-based temporal self-similarity method groups body parts of a hierarchical model into several sets and calculates movements for each set. While recognition results can depend on how the sets are made, the best way to find optimal sets is to separate frequently used body parts from less-used body parts. Then, we apply a multiclass support vector machine whose optimization algorithm is based on structural support vector machines. In this paper, the effectiveness of the proposed feature extraction algorithm is demonstrated in an application for taebo gesture recognition. We show that the model-based temporal self-similarity method can overcome the shortcomings of the conventional temporal self-similarity method and the recognition results of the model-based method are superior to that of the conventional method.