• 통상정보연구
• /
• 제6권3호
• /
• pp.159-180
• /
• 2004

This paper focus on building online trust in electronic commerce between partner that have never traded with each other before, the so-called first trade situation. For this, this paper proposes the model to build trust for conduction first trade transaction in e-trade(so-ca1led Trust Matrix Model : TMM). The TMM is based on the idea that for business to business electronic trade a balance has to be found between anonymous procedural trust, i.e. procedural solutions for trust building, and personal trust based on positive past experiences within for first trade situation, because of the lack of experience in these situations. The procedural trust solutions are related in the notion of institution-based trust, because the trust in the procedural solutions depends on the trust one has in the institutions that issued or enforces the procedure. The TMM can be used as a tool to analyze and develope trust-building services to help organizations conduct first -trade electronic trade.

• 대한수학회지
• /
• 제54권2호
• /
• pp.613-626
• /
• 2017

For the overdetermined linear system, when both the data matrix and the observed data are contaminated by noise, Total Least Squares method is an appropriate approach. Since an ill-conditioned data matrix with noise causes a large perturbation in the solution, some kind of regularization technique is required to filter out such noise. In this paper, we consider a Dual regularized Total Least Squares problem. Unlike the Tikhonov regularization which constrains the size of the solution, a Dual regularized Total Least Squares problem considers two constraints; one constrains the size of the error in the data matrix, the other constrains the size of the error in the observed data. Our method derives two nonlinear equations to construct the iterative method. However, since the Jacobian matrix of two nonlinear equations is not guaranteed to be nonsingular, we adopt a trust-region based iteration method to obtain the solution.

• 정보보호학회논문지
• /
• 제28권1호
• /
• pp.241-248
• /
• 2018

추천시스템(Recommender System, RS)는 정보 과잉 공급 상태에서 사용자들에게 최적화된 정보를 제공하는 시스템이다. RS의 핵심은 사용자의 행동 결과를 정확하게 예측하는 것이다. 이러한 예측을 위해 Matrix Factorization (MF) 방식이 초기에 사용되었으며, 최근 SNS의 발달에 따라 Social Information을 추가적으로 활용하여 예측 정확도를 높이고 있다. 본 논문에서는 기존 연구에서 간과 되었던 RS 내부 trust cluster를 이용하여 추가적으로 성능을 향상시키고, trust cluster의 특성에 대하여 분석한다. 기존 방법론 3가지와 비교한 결과 본 논문에서 제안하는 방식이 가장 높은 정확도를 보임을 확인하였다.

• Journal of applied mathematics & informatics
• /
• 제18권1_2호
• /
• pp.1-23
• /
• 2005

The decentralized static output feedback design problem is considered. A constrained trust region method is developed that solves this optimal control problem when a complete set of state variables is not available. The considered problem is interpreted as a non-linear (non-convex) constrained matrix optimization problem. Then, a decentralized constrained trust region method is developed for this problem class exploiting the diagonal structure of the problem and using inexact computations. Finally, numerical results are given for the proposed method.

• Archives of Plastic Surgery
• /
• 제47권6호
• /
• pp.567-573
• /
• 2020

Background Polyurethane implants have been used on and off in breast reconstruction since 1991 while prepectoral breast reconstruction has gained popularity in recent times. In this study, we present our outcomes from the use of acellular dermal matrix (ADM) complete wrap with polyurethane implants in prepectoral breast reconstruction. Methods This is a retrospective review of prospectively maintained database from 41 patients receiving complete ADM wrap with prepectoral polyurethane implants over a 3-year period. Selection criteria were adapted from a previous study (4135 Trust Clinical Audit Database) evaluating prepectoral reconstruction with Braxon matrices. Patient demographics, operative data, surgical complications, and outcomes were collected and analyzed. Results A total of 52 implant reconstructions were performed in 41 patients with a mean follow-up of 14.3 months (range, 6-36 months). The overall reported complication rates including early (less than 6 weeks) and late complications. Early complications included two patients (4.9%) with wound dehiscence. One of which had an implant loss that was salvageable. Another patient (2%) developed red-breast syndrome and two women (4.9%) developed with seroma treated conservatively. Late complications included one patient (2%) with grade II capsular contraction, 12 patients with grade I-II rippling and two patients (4.9%) with grade III rippling. Conclusions We present our experience of prepectoral polyurethane implant using complete ADM wrap. This is one of the few papers to report on the outcome of the prepectoral use of polyurethane in immediate implant-based breast reconstruction. Our early observational series show satisfactory outcome and long-term results are warranted by a large multicenter study.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권5호
• /
• pp.2082-2102
• /
• 2018

In recent years, social network related applications such as WeChat, Facebook, Twitter and so on, have attracted hundreds of millions of people to share their experience, plan or organize, and attend social events with friends. In these operations, plenty of valuable information is accumulated, which makes an innovative approach to explore users' preference and overcome challenges in traditional recommender systems. Based on the study of the existing social network recommendation methods, we find there is an abundant information that can be incorporated into probability matrix factorization (PMF) model to handle challenges such as data sparsity in many recommender systems. Therefore, the research put forward a unified social network recommendation framework that combine tags, trust between users, ratings with PMF. The uniformed method is based on three existing recommendation models (SoRecUser, SoRecItem and SoRec), and the complexity analysis indicates that our approach has good effectiveness and can be applied to large-scale datasets. Furthermore, experimental results on publicly available Last.fm dataset show that our method outperforms the existing state-of-art social network recommendation approaches, measured by MAE and MRSE in different data sparse conditions.

• The Journal of Asian Finance, Economics and Business
• /
• 제8권5호
• /
• pp.961-969
• /
• 2021

The study aimed to investigate the level of customer trust, loyalty, and re-purchase intention of coastal tourism in Thailand during the COVID-19 crisis; to test the different levels of customer trust, loyalty, and re-purchase intention by local tourists between East-side and West-side coasts of Thailand; and to examine the influence of customer trust and loyalty on re-purchase intention in coastal tourism. Using multistage sampling, this study sampled 487 Thai local tourists who had experienced coastal tourism in Thailand during the COVID-19 crisis. A questionnaire, descriptive analysis, independent sample t-test, correlation matrix, and multiple regression analysis were used to collect and analyze the data. All customer trust, loyalty, and re-purchase intentions in coastal tourism by local tourists were at a high level. There were significantly different levels of customer trust, loyalty, and re-purchase intentions by local Thai tourists between the East-side and West-side coasts of Thailand. Moreover, the study found that there was a significant positive influence of customer trust and loyalty on re-purchase intentions in coastal tourism by local tourists in Thailand during the COVID-19 crisis. This study indicates that Thai tourism industry still must develop and improve its local customer loyalty and trust because these positively influence customer re-purchase intentions.

• Journal of Mechanical Science and Technology
• /
• 제16권2호
• /
• pp.203-210
• /
• 2002

For effective response surface modeling during sequential approximate optimization (SAO), the normalized and the augmented D-optimality criteria are presented. The normalized D-optimality criterion uses the normalized Fisher information matrix by its diagonal terms in order to obtain a balance among the linear-order and higher-order terms. Then, it is augmented to directly include other experimental designs or the pre-sampled designs. This augmentation enables the trust region managed sequential approximate optimization to directly use the pre-sampled designs in the overlapped trust regions in constructing the new response surface models. In order to show the effectiveness of the normalized and the augmented D-optimality criteria, following two comparisons are performed. First, the information surface of the normalized D-optimal design is compared with those of the original D-optimal design. Second, a trust-region managed sequential approximate optimizer having three D-optimal designs is developed and three design problems are solved. These comparisons show that the normalized D-optimal design gives more rotatable designs than the original D-optimal design, and the augmented D-optimal design can reduce the number of analyses by 30% - 40% than the original D-optimal design.

• 지능정보연구
• /
• 제22권3호
• /
• pp.113-127
• /
• 2016

협업 필터링(Collaborative Filtering)은 유용성과 정교성 면에서 가장 성공적인 추천 알고리즘으로 평가받으며 산업계나 학계에서 많이 활용 및 연구되고 있지만, 기본적으로 사용자들이 평가한 점수에만 기반하여 추천결과를 생성하는 한계점이 있다. 이에 본 연구는 사용자가 상품을 구매할 때 자신이 신뢰하는 타인의 추천을 더 적극적으로 수용할 것이라는 점에 착안하여, 사용자의 평점 외에 사용자 간 신뢰관계를 소셜네트워크분석으로 분석한 결과를 추가로 반영하는 추천 알고리즘들을 제안하였다. 구체적으로 본 연구에서는 소셜네트워크분석에서 네트워크 내의 중심적 위치를 나타내는 척도인 내향 및 외향 중심성을 활용하여 사용자 간 유사도를 산출하는 알고리즘들과 사용자 신뢰 네트워크를 탐색하여 추천 대상이 되는 사용자가 직접 간접적으로 신뢰하는 사용자의 평가점수를 보다 높게 반영하는 알고리즘을 제안한 뒤 그 성능을 비교해 보았다. 실제 데이터에 적용하여 분석한 결과, 사용자 신뢰 네트워크의 내향 중심성 지수를 조건 없이 적용한 경우에는 오히려 정확도의 감소만을 야기하는 것으로 나타났고, 일정 임계치 이상의 외향 중심성을 갖는 사용자에 한해 내향 중심성 지수를 고려한 추천 알고리즘은 전통적인 협업 필터링에 비해 약간의 정확도 개선이 이루어짐을 확인할 수 있었다. 아울러, 사용자 신뢰 네트워크를 기반으로 탐색하는 알고리즘이 가장 우수한 성능을 보이는 것을 알 수 있었으며, 전통적인 협업 필터링과 비교해서도 통계적으로 유의한 수준의 정확도의 개선이 이루어짐을 확인할 수 있었다.

• 디지털융복합연구
• /
• 제15권7호
• /
• pp.467-477
• /
• 2017

본 연구는 IPA를 통해 수영복 소비자들이 인식하고 있는 수영복 브랜드이미지에 대한 중요도와 만족도에 대한 순위와 차이를 제시하고 IPA 매트릭스로 도식화하여 수영복 산업시장의 활성화를 위한 시사점을 제공하는데 연구의 목적이 있다. 이를 위해 대학생 및 대학원생들을 대상으로 298명의 설문조사 자료를 분석하였고, 평균값으로 중요도와 만족도의 순위와 대응표본 t-test, IPA매트릭스 도식화를 통해 다음과 같은 결론을 도출하였다. 첫째, 수영복 브랜드이미지에 대한 중요도와 만족도의 평균값을 통해 순위를 살펴본 결과, 중요도에서 '품질', '기능성', '가격', '디자인', '신뢰감', '색상' 순으로 평균 4.0이상으로 나타났고 만족도에서 '신뢰감', '품질', '기능성', '인지도' 순으로 평균 3.5이상으로 나타났다. 둘째, 수영복 브랜드이미지에 대한 차이검증 결과 '품질', '가격', '기능성', '디자인', '색상', '신뢰감', '세련미' 순으로 유의미한 차이를 보였다. 반면, 인지적 이미지는 전혀 유의미한 차이를 보이지 않았다. 셋째, 수영복 브랜드이미지에 대한 IPA결과 I사분면(현상유지)은 '디자인', '기능성', '품질', '신뢰감'요인이 II사분면(집중노력지향)은 '가격'과 '색상'요인이 III사분면(열등순위)은 '광고이미지', '이벤트', '유행성', '독창성'요인이 IV사분면(과잉노력지양)은 '인지도', '세련미'요인이 포함 되어있었다.