• Proceedings of the Korea Information Processing Society Conference
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.293-294
• /
• 2023

이 연구는 사람의 외적인 요소와 마이어스-브릭스가 제안한 16 가지 성격유형을 결부한 연구이다. 기존의 성격유형에 대한 기준을 기반으로 하여, 일반적인 MBTI 판독기와는 다르게 StyleCLIP 을 활용해서 추상적인 단어로 이미지를 변환하고 전이학습 AI 를 이용하여 비교 테스트를 진행한다. 최종적으로 이 연구를 통해 외모와 성격은 연관이 있다는 가설을 증명한다.

• Journal of Korean Academy of Fundamentals of Nursing
• /
• 제22권3호
• /
• pp.297-307
• /
• 2015

Purpose: The purpose of this study was to identify the learning style, learning efficacy, transfer of learning, and confidence in performance competence of Core Basic Nursing Skills (CBNS) and factors influencing confidence in performance competence of CBNS by nursing students. Methods: A descriptive study design was used. Participants were 148 nursing students. Data were analyzed using SPSS 20.0 for descriptive statistics, ANCOVA, correlation and regression. Results: Learning styles of the participants were assimilator 33.11%, accommodator 26.35%, diverger 23.65%, and converger 16.89%. Learning efficacy was significantly different according to learning styles, however, transfer of learning and confidence in performance competence of CBNS were not significantly different according to learning styles. Confidence in performance competence of CBNS positively correlated with transfer of learning and learning efficacy. Transfer of learning was a significant predictor of confidence in performance competence of CBNS. Conclusion: The findings of this study indicate that transfer of learning influences confidence in performance competence of CBNS. Thus, nursing faculty should develop educational strategies to enhance and improve transfer of learning, and development of effective confidence in performance competence of CBNS programs.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제10권9호
• /
• pp.4568-4584
• /
• 2016

Steganography based on text generation has become a hot research topic in recent years. However, current text-generation methods which generate texts of normal style have either semantic or syntactic flaws. Note that texts of special genre, such as poem, have much simpler language model, less grammar rules, and lower demand for naturalness. Motivated by this observation, in this paper, we propose a text steganography that utilizes Markov chain model to generate Ci-poetry, a classic Chinese poem style. Since all Ci poems have fixed tone patterns, the generation process is to select proper words based on a chosen tone pattern. Markov chain model can obtain a state transfer matrix which simulates the language model of Ci-poetry by learning from a given corpus. To begin with an initial word, we can hide secret message when we use the state transfer matrix to choose a next word, and iterating until the end of the whole Ci poem. Extensive experiments are conducted and both machine and human evaluation results show that our method can generate Ci-poetry with higher naturalness than former researches and achieve competitive embedding rate.

• Annual Conference on Human and Language Technology
• /
• 한국정보과학회언어공학연구회 2018년도 제30회 한글 및 한국어 정보처리 학술대회
• /
• pp.434-439
• /
• 2018

문체 변환 시스템을 학습하는 데 있어서 가장 큰 어려움 중 하나는 병렬 말뭉치가 부족하다는 것이다. 최근 대량의 비병렬 말뭉치만으로 문체 변환 문제를 해결하려는 많은 연구들이 발표되었지만, 아직까지도 원 문장의 정보 보존(Content preservation)과 문체 변환(Style transfer) 모두를 이루는 것이 쉽지 않은 상태이다. 특히 비교사 학습의 특성상 문체 변환과 동시에 정보를 보존하는 것이 매우 어렵다. Attention 기반의 Seq2seq 네트워크를 이용할 경우에는 과도하게 원문의 정보가 보존되어 문체 변환 능력이 떨어지기도 한다. 그리고 OOV(Out-Of-Vocabulary) 문제 또한 존재한다. 본 논문에서는 Attention 기반의 Seq2seq 네트워크를 이용하여 어절 단위의 정보 보존력을 최대한 높이면서도, 입력 문장에 효과적으로 Noise를 넣어 문체 변환 성능을 저해하는 과도한 정보 보존 현상을 막고 문체의 특성을 나타내는 어절들이 잘 변환되도록 할 뿐 아니라 OOV 문제도 줄일 수 있는 방법을 제안한다. 우리는 비교 실험을 통해 본 논문에서 제안한 방법들이 한국어 문장뿐 아니라 영어 문장에 대해서도 state-of-the-art 시스템들에 비해 향상된 성능을 보여준다는 사실을 확인하였다.

• Fashion & Textile Research Journal
• /
• 제8권5호
• /
• pp.562-570
• /
• 2006

Knit industry requires versatile knit designers who have to control whole process such as material select, color, type of machine, technique, structure and shape. In other words, knit designers urgently need to interplay between technology and design aesthetics. It becomes essential to learn and master basic principles of knitting in order to design and develop effectively for innovative knitwear design. The main purpose of this study is to show fresh methodology through examining the various applications of 'Stitch transfer knitting structure' used in women's knitwear design development. Stitch transfer knitting structure is one of the most popular techniques among various basic knitting methods. Eight knitwears were designed and produced with various types of formative applications which differ from existent stitch transfer knitwears of flat and simple style in the market. These are produced by computer knitting machine 'Shima Seiki 122S' and programmed by 'Knit CAD' software. Among various basic stitch transfer pattern, 'Leaf' pattern is selected as a most effective example for this study. This fabric applied to collar, sleeves, body pattern and accessories as formative decoration. It is expected that this study of methodology helps to get open mind and to indicate possibility to develop and show creative and innovative knitwear and knitted textile design.

• Journal of the Korea Fashion and Costume Design Association
• /
• 제12권4호
• /
• pp.207-221
• /
• 2010

This study aim to suggest how alter the korean traditional image design for umbrellas by sublimation transfer digital textile printing. Umbrellas are highly depend on design and made from polyester fiber which is proper item to utilize sublimation transfer digital textile printing. Sublimation transfer digital textile printing system can use computer system to create the delicate high dense images and one full layout through the hole umbrella. It can create distinctive style of design compare with former screen printing umbrella design. As a result of this study, Korean traditional images were adopted and recreated for umbrellas as the modern practical item. 7 of umbrella designs were developed and sample umbrellas could be produced in short period comparing with screen printing process. Through this study, as green printing process, sublimation transfer digital textile printing will be more applied to manufacture high quality textile products along with design development, thus it is expected as an alternative plan to leads growth of umbrella industries.

• Journal of Energy Engineering
• /
• 제24권1호
• /
• pp.24-31
• /
• 2015

The purpose of this study is to provide the thermal comfort range according to the residential style and clothing level at radiant floor heating space, and compare the annual energy consumption and energy cost for each condition. Lower neutral point temperature has been stood for floor sitting style than chair sitting style, which appears that the thermal sensation was affected by local heat transfer between floor surface and the human body. The result of research indicates that neutral point temperature was in inverse proportion with the clothing level. It is interpreted that the increasing of clothing level results decrement of heat loss from human body, and is available to achieve same thermal comfort at lower room temperature. It was analyzed that the floor sitting style is more economical residential style than the chair sitting style, because the energy consumption of the floor sitting style is saved by 6.0% in average to compare with that of the chair sitting style. It is analyzed that energy consumption has been decreased by 13.5% with the clothing level of 1.2 Clo than with that of 1.0 Clo, and decreased by 18.0% than with that of 0.8 Clo, which explains that the energy saving can be achieved with the variation in life habit to increase the clothing level.

• Bulletin of the Korean Chemical Society
• /
• 제28권8호
• /
• pp.1358-1362
• /
• 2007

The structures, energetics and transfer integrals of the acene tetramers up to pentacene are investigated with the ab initio molecular orbital method at the level of second-order Møller-Plesset perturbation theory (MP2). Calculated geometries for the herringbone-style structures found in the crystal structure were characterized as local minima, however the geometrical discrepancy between crystal and MP2 theoretical structure is reasonably small. The binding energy of pentacene tetramer was calculated up to 40 kcal/mol (MP2/6-31G(d)) and about 90 kcal/mol (MP2/aug-cc-pVDZ), and the latter seems to be too much overestimated. The tendency of the hole transfer integrals computed with ab initio MP2/3-21G(d) geometry is well agreement with those estimated with crystal structure with some discrepancy, and the gradual increment of the transfer integrals at the crystal geometry is attributed to mainly packing structure rather than the intrinsic property of acene such as a size of acene.

• 한국신재생에너지학회:학술대회논문집
• /
• 한국신재생에너지학회 2006년도 춘계학술대회
• /
• pp.110-112
• /
• 2006

Membrane electrode assembly (MEA) for polymer electrolyte fuel cell (PEFC) are commonly prepared in the research laboratory by spraying, screen-printing and brushing catalyst slurry onto membrane or other support material like carbon paper or polyimide film in a batch style. These hand applications of the catalyst slurry are painstaking process with respect to precision of catalyst loading and reproducibility. It has been generally mentioned that the adoption of continuous process is very helpful to develop the reliable product. In the present work, we report the results of using continuous type coater with doctor-blade to coat catalyst slurry for preparing the MEA catalyst layers In a faster and highly reproducible fashion. We show that while expectedly faster than batch style, the machine coater requires the use of slurry of appropriate composition and a properly selected transfer decal material in order to achieve superior MEA plat lnw loading reproducibility. To make highly viscous catalyst slurry that is imperative for using coater, we use 40wt.% Nafion solution and minimize the content of organic solvent. And the choice of proper high surface area catalyst is important in the viewpoint of making well-dispersed slurry. After catalyst coating onto the support material, we transferred the catalyst layer to both sides of Nafion membrane by hot-pressing In this case, the degree of transfer was Influenced by hot-pressing condition including temperature, pressure, and time. To compare the transferring ability, we compared so many films and detaching papers. And among the support, polyethylene terephthalate(PET) film shows the prominent result.

• The Journal of Society for e-Business Studies
• /
• 제26권1호
• /
• pp.29-41
• /
• 2021

Recently, there are increasing attempts to utilize deep learning methodology in the fashion industry. Accordingly, research dealing with various fashion-related problems have been proposed, and superior performances have been achieved. However, the studies for fashion style classification have not reflected the characteristics of the fashion style that one outfit can include multiple styles simultaneously. Therefore, we aim to solve the multi-label classification problem by utilizing the dependencies between the styles. A multi-label recognition model based on a graph convolution network is applied to detect and explore fashion styles' dependencies. Furthermore, we accelerate model training and improve the model's performance through transfer learning. The proposed model was verified by a dataset collected from social network services and outperformed baselines.