• 한국컴퓨터정보학회논문지
• /
• 제29권1호
• /
• pp.31-40
• /
• 2024

국내 온라인 패션 플랫폼은 개인사업자가 제품정보를 직접 등록하기 때문에 개인사업자의 불편함을 초래한다. 많은 제품군을 한꺼번에 수동 등록하므로 수기 입력된 제품정보로 인한 신뢰성 문제가 발생한다. 등록된 상품 이미지의 저품질 및 데이터 수의 불균형으로 인한 편향도 심각하게 제기된다. 본 연구는 오버샘플링 기법을 통해 데이터 편향을 최소화하고 13개 패션 카테고리의 다중 분류를 수행하는 ResNet50 모델을 제안한다. 컴퓨팅 자원과 오랜 학습시간을 최소화하기 위해 전이학습을 활용했다. 결과적으로, 데이터 수가 매우 부족했던 클래스의 데이터 증강을 통해 기본 CNN 모델에 비해 최대 33.4%의 향상된 식별력을 보여주었다. 모든 결과의 신뢰성은 정밀도-재현율 곡선으로 보장한다. 본 연구는 국내 온라인 패션 플랫폼 산업의 발전을 한 단계 끌어올릴 수 있을 것으로 기대한다.

• 정보처리학회논문지B
• /
• 제17B권1호
• /
• pp.93-98
• /
• 2010

본 논문에서는 폭발적으로 증가하고 있는 상품평을 효과적으로 활용하기 위해 언어적 분석을 통하여 상품 평가를 요약하는 시스템을 제안한다. 시스템에서는 스커트 상품 분류의 경우 '디자인'과 '재질'과 같이, 상품을 평가하는 기준이 되는 항목에 대한 상품평의 부정과 긍정의 극성 여부를 판별하여 그래프 형태로 요약하여 제시한다. 본 논문에서는 작은 시드 어휘와 문맥에 기반한 자동 확장 방법을 사용하여 평가 항목 별 평가 어휘 극성 사전을 구축하여 평가 항목에 대한 상품평의 극성을 판정한다. 제안한 방식은 여러 온라인 쇼핑몰의 실제 상품평에 대한 실험에서 극성 사전 추출에서 평균 69.8%의 정확율과 문장별 극성 식별에서 평균 81.8%의 정확율을 보였다.

• 호남수학학술지
• /
• 제42권2호
• /
• pp.219-234
• /
• 2020

The fundamental relation on a fuzzy hyperring is defined as the smallest equivalence relation, such that the quotient would be the ring, that is not commutative necessarily. In this paper, we introduce a new fuzzy strongly regular equivalence on fuzzy hyperrings, where the ring is commutative with respect to both sum and product. With considering this relation on fuzzy hyperring, the set of the quotient is a commutative ring. Also, we introduce fundamental functor between the category of fuzzy hyperrings and category of commutative rings and some related properties. Eventually, we introduce α-part in fuzzy hyperring and determine some necessary and sufficient conditions so that the relation α is transitive.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권9호
• /
• pp.3680-3692
• /
• 2020

The purpose of this study is to orient users' touchpoints in co-design experience, to identify their need via visualized experience map, to recommend valid design information in online e-customization services. A user-centered co-design experience map (UCEM) is adopted to analyze the relation between users' desire and time spent, so as to evaluate the online co-design experiences. Based on evolutionary algorithm and fuzzy theory, data of this study is collected from 30 participants. The data was analyzed by descriptive analysis in SPSS, and frequency query and word cloud in NVivo. Employing design category and evaluating users' time spent, the findings are that (a) vamp color matching is consistent with interview data; (b) supported by qualitative feedback, the virtual experience map played an important role in the co-design process and the visualized interaction process; and (c) participants prefer to get more information and professional help on color matching and exterior design. Based on the findings in design category, future work should be focused on developing a better understanding of design resource recommendations and multi-stakeholder communication.

• Asia Marketing Journal
• /
• 제17권1호
• /
• pp.55-67
• /
• 2015

Despite the well-known predictive power of Dirichlet model on customer loyalty, deviations of share of category requirement (SCR) predicted by Dirichlet model from actual SCR have been repeatedly reported. It has been shown that these deviations can be systematically explained by some factors such as brand's market share, product positioning strategy, purchase volume and retail marketing mix strategies. Presuming that brand equity would be additional sources of these deviations, current study assesses the incremental predictive power of brand equity by using over 4,000 brand-level observations for the consumer packaged goods industry in the U.S. Our model estimations indicate that brands that exhibit higher brand equity enjoy excess loyalty, with the primary driver being volume premium, rather than price premium. Overall, our findings support the notion that idiosyncratic brand properties can explain excess behavioral loyalty, a notion that is in stark contrast with the Dirichlet view of the world: brand equity does not exist.

• 반도체디스플레이기술학회지
• /
• 제22권2호
• /
• pp.40-44
• /
• 2023

The defect detection based on deep learning requires minimal loss and high accuracy to pinpoint product defects. In this paper, we confirm the loss rate of deep learning training based on disc-shaped artificial scaffold images. It is intended to compare the performance of Cross-Entropy functions used in object detection algorithms. The model was constructed using normal, defective artificial scaffold images and category cross entropy and sparse category cross entropy. The data was repeatedly learned five times using each loss function. The average loss rate, average accuracy, final loss rate, and final accuracy according to the loss function were confirmed.

• Environmental Analysis Health and Toxicology
• /
• 제19권2호
• /
• pp.227-233
• /
• 2004

The main objectives of this study were to identify from literature review the potential sources and to provide a preliminary national emission inventory for the unintentionally produced polychlorinated biphenyls (PCBs) (i.e., by - product PCBs). In Korea, fuel combustion, waste combustion, thermal industrial processes, and transportation were identified as potential sources of by -product PCB s. According to the availability of the emission factors and/or activity data, emission inventory could be assessed only for fuel combustion, waste combustion, steel industry, non-ferrous industry, and non-metallurgical industry. The total national emission of by-product PCBs was estimated to be 1087kg for the year 2000. The preliminary estimation further indicated that the steel manufacturing was the single dominant emission category, contributing 93％ to the total emission. Of the steel manufacturing processes, the contribution of the electric arc furnace was about 80％ of the total emission. Due to high uncertainty associated with both the emission factors and activity statistics, the emission estimates in this study are likely to contain significant errors. However, the results of the present work could serve the first step toward future efforts to establish national source and emission inventories of by-product PCBs.

• 복식문화연구
• /
• 제13권6호통권59호
• /
• pp.934-945
• /
• 2005

The purpose of this study was to investigate the content and presentation style of consumer information of Korean young casual apparel brand. To collect the data for this study, the representative 25 casual brand web sites were selected, based on major search engines. In addition, to investigate exact product information, four product categories, knit shirts and casual pants for men and women, were selected and the number of products was limited as maximum 15 products per each category. A coding instrument was developed to capture the consumer information, based on the instrument by Park and Stoel(2002). The Pretest was conducted to gauge inter-coder reliability and the results showed that inter-coder reliability was highly acceptable. The results of this study were as follows. Most casual brand web sites for this study were presented well in brand and customer service information. Especially, many web sites provided various engaging information such as various events(best dresser contest, date with a star, special gift) and useful multimedia file(MP3 music file, screen saver, movie, calender). However, product information was very lack in most web sites. Especially, sizing and fitting information and textile and fabric hand information were rarely provided. Therefore, this result showed that the web sites should provide more specific product information and develop devices to get tactile sensory and experiential information for enhancement of future e-commerce of apparel products.

• 한국인쇄학회지
• /
• 제29권3호
• /
• pp.31-40
• /
• 2011

Using of QR(Quick Response) code is dramatically extended to various marketing area; not only substitute of bar code but also new tool of PR and marketing. This research attempts to analyze brand communication effect using QR code printed on product package especially in snack product category. Findings are 1) communication effect are different according to the type of book-trailers; 2) cinematic production and animation are the most effective type of book-trailers; 3) for memory and confirmation(sharing), a)stills and straplines, Analysis result of hypothesis I showed that characters of QR code influence on brand attitude. Among dependent variables, only reliability is significant. That means reliability of company and brand using QR code influence on brand attitude. The higher reliability of QR code, the better brand attitude of the brand. Analysis result of hypothesis II found that only reliability is significant on purchasing intention. Reliability of company and product using QR code influences on purchasing intention. The higher reliability of QR code, the higher possibility of purchasing products. Therefore, company can enhance reliability of both company and its products by using QR code. Using QR code will bring high reliability and high brand attitude and purchasing intention.

• 한국의류학회지
• /
• 제27권9_10호
• /
• pp.1036-1047
• /
• 2003

This study was carried out in the purpose of proposing internet marketing strategy which can make conversion rate higher through analysis of internet shopping behavior types. This study was executed in two stages; qualitative study, quantitative study. In the qualitative study, internet shopping behavior types were investigated through the In-depth interview and direct observation, and then in the quantitative study, differences of internet shopping behavior types according to consumer characteristics, product properties and shopping-mall types were examined. For qualitative study, 30 samples by focus sampling were inquired and for quantitative study, 334 data were collected through web survey. The results of this study are as follows: First, 7 Internet shopping behavior types of apparel were found through the qualitative study: cautious purchase by price comparison, searching purchase, special low price purchase, impulse purchase, prepurchase deliberation, information accumulation, recreation-oriented. Second, in relation to consumer characteristics, consumers that have many internet purchase experiences showed goal-directed behavior more and female did more special low price purchase behavior and impulse purchase behavior than male. Third, according to product properties, high price product led more cautious purchase by price comparison & prepurchase deliberation behavior and fashionable product led more information-searching behavior. In the case of low price and fashionable products, impulse purchase behaviors were showed more. Forth, according to Internet shopping mall types, category killer shopping mall visitors showed information search behavior, recreation-oriented behavior more.