• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• 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.

• Asia Marketing Journal
• /
• v.19 no.1
• /
• pp.1-17
• /
• 2017

Cause-related Marketing (CrM) has become an increasingly popular marketing approach over the past two decades. However, neither researchers nor organizations fully understand the determinants of a successful CrM partnership. This research fills this gap. Specifically, we employ the schema theory to explore circumstances in which the CrM alliance cannot achieve a success. We use a theoretical modeling approach to report that, when consumers' typicality-based cognitive process is assumed, the CrM activity with the partners' more-discrepant attribute profile cannot be evaluated favorably, but the attribute-level uncertainty about the CrM alliance is less likely to feedback to the two partners. Furthermore, we argue that, under the schema-plus-tag model, consumers may not like the CrM program with a similar attribute profile. Therefore, this CrM approach may fail. To our knowledge, we are the first to apply the schema theory to explain how a CrM alliance can achieve a success.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1998.10a
• /
• pp.103-106
• /
• 1998

본 연구는 연구·개발 프로젝트의 평가를 위하여 연구·개발 프로젝트의 다-속성(Multi-attributes)을 고려하한 평가 모델의 연구이다. 이를 위하여 우선 평가구조를 구축하고 각 속성별 평가를 종합하기 위한 종합성과도(Total Preference Index)로 단일 측정치로 평가 할 수 있도록 종합하기 위한 적절한 효용함수를 도입하여 사용하였다. 이러한 평가 과정을 다-속성 의사결정 모델(Multi-attribute Utility Model)로 통합하였으며 연구·개발프로젝트의 특성을 고려하여 각 연구실의 책임자(Laboratory Directors)의 평가체계를 개발하여 본 평가모델에 포함하였다. 본 평가모델의 시험 적용을 위하여 특정 연구소에 시험적용하고 그 결과를 보였으며, 부분적으로 보완 연구될 경우 일반적인 프로젝트의 평가모델로 활용될 수 있으리라 생각된다.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.279-284
• /
• 2017

The visual impact of fonts on lots of documents and design work is significant. Accordingly, the users desire to appropriately use fonts suitable for their intention. However, existing font recommendation programs are difficult to consider what users want. Therefore, we propose a font recommendation system based on user-evaluated font attribute value. The properties of a font are called attributes. In this paper, we propose a font recommendation module that recommends a user 's desired font using the attributes of the font. In addition, we classify each attribute into three types of usage, personality, and shape, suggesting the font that is closest to the desired font, and suggest an optimal font recommendation algorithm. In addition, weights can be set for each use, personality, and shape category to increase the weight of each category, and when a weight is used, a more suitable font can be recommended to the user.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.44-49
• /
• 2001

This paper is concerned with development of a multi-attribute structured decision model. In this study, we used AHP(analytic hierarchy process) and fuzzy set ranking methodology to overcome the multi-attributes structured decision problems ; such as multi-objective, multi-criterion, and multi-attributes. We proposed a 2-step approach : 1) individual evaluation and 2) integration of individual evaluations. In the first step, we define the performance factors and construct ana]isis structure, and in the second step performance evaluation by individual evaluators, and in second step, the results of individual evaluations are integrate. Also we developed a systematic and practical computer program to solve the problems according to the proposed methods. The proposed approach was known to be effective through a set of sample problems.

• Journal of the Korean housing association
• /
• v.12 no.3
• /
• pp.109-116
• /
• 2001

Street is an important factor that street quality determines urban quality as the space of the high public nature in city. The research method considering sequence landscape is important because it is considered that the impression of such a street is formed by the wholeness of consecutive landscape and experience in moving street. Therefore, this study are pursued to structure the analysis of psychological evaluation and environmental attribute base on the physical characteristics of sequence landscape to supply the basic materials for seeking the national direction as qualitative improvement of urban street space and the attractive urban street space. Then the results are as follows. It is shown that the psychological evaluation structure of sequence landscape is evaluated by ornamental nature, intimacy, orderliness, openness and movability and that of environmental attributes is evaluated by four factors such as activity, intimacy, spatial attribute and openness.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06c
• /
• pp.166-168
• /
• 2012

바이오 문헌으로부터의 의미 있는 객체 추출 및 상호작용 관계 추출은 수 많은 바이오 문헌으로부터 유용한 정보를 얻기 위한 필수적인 과정이다. 특히 문헌으로부터 유전자 또는 단백질 이름과 같은 바이오 객체를 정확하게 인지하는 것은 새로운 객체인식의 어려움과 객체를 찾기 위한 특징 패턴의 다양성으로 인해 도전적인 과제로 남아있다. 본 논문에서는 전처리 과정을 거친 문헌 데이터로부터 12개의 의미 있는 속성들을 선택하였다. 선택된 속성에 데이터마이닝 기법중 하나인 속성 추출 기법을 적용하여 객체를 분류하는데 있어 의미 있는 속성들을 추출하였다. 특징 추출 방법과 분류 알고리즘이 분류 성능에 미치는 영향을 평가하기 위해 각 방법의 정확도를 사용하여 분류 성능을 비교였으며, Gain Ratio Attribute Evaluation과 Symmetrical Uncertainty Attribute Evaluation 기법에 의해 추출된 속성이 가장 정확한 분류 성능을 보여주었다.

• The Journal of Society for e-Business Studies
• /
• v.14 no.3
• /
• pp.15-37
• /
• 2009

This study presents a new retrieval system developed to consider various preferential requirements for buyers in the real estate market. The paper analyses essential factors affecting the price of real estate and then a set of factors are classified by region-related factor and individual-related factor. After endowing the buyer's selected factors with weights in the retrieval system, the optimal solutions have been drawn by comparing with the others through an entropy measure of Multi-attribute Decision Making. This retrieval system is applied to the Busan real estate market to estimate the solutions of retrieval. Evaluation results indicate that the retrieval system can provide useful information to analyse the price determination factors of real estate, as well as to save the searching cost of the buyers.

• Journal of the Korea Convergence Society
• /
• v.7 no.5
• /
• pp.181-188
• /
• 2016

This study investigated consumer convergence preferences for healthcare information service in social media using the multi-attribute utility theory. The multi-attribute utility theory is an approach for compositional modeling attributes wherein researchers compute the overall service utility by aggregating the evaluation results for attribute values. We found that healthcare information service consumers in social media evaluate the service preference through the multi attributes - Update, Cost, Usefulness, Usability, Reliability. And the study shows that individual preference by estimating weight of each attribute in the services influences the service use intention. So we believe that estimating preferences with multi-attribute utility theory will predict business success of the services in a new media, and for a successful business of the social media services marketers should assess the consumer's preference using the multi-attributes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.11
• /
• pp.5229-5252
• /
• 2016

With proliferation of diverse network access technologies, users demands are also increasing and service providers are offering a Quality of Service (QoS) to satisfy their customers. In roaming, a mobile node (MN) traverses number of available networks in the heterogeneous wireless networks environment and a single operator is not capable to fulfill the demands of user. It is crucial task for MN for selecting a best network from the list of networks at any time anywhere. A MN undergoes a network selection situation frequently when it is becoming away from the home network. Multiple Attribute Group Decision (MAGD) method will be one of the best ways for selecting target network in heterogeneous wireless networks (4G). MAGD network selection process is predominantly dependent on two steps, i.e., attribute weight, decision maker's (DM's) weight and aggregation of opinion of DMs. This paper proposes Multi-Attribute Intuitionistic Fuzzy Group Decision Method (MAIFGDM) using TOPSIS for the selection of the suitable candidate network. It is scalable and is able to handle any number of networks with large set of attributes. This is a method of lower complexity and is useful for real time applications. It gives more accurate result because it uses Intuitionistic Fuzzy Sets (IFS) with an additional parameter intuitionistic fuzzy index or hesitant degree. MAIFGDM is simulated in MATLAB for its evaluation. A comparative study of MAIFDGM is also made with TOPSIS and Fuzzy-TOPSIS in respect to decision delay. It is observed that MAIFDGM have low values of decision time in comparison to TOPSIS and Fuzzy-TOPSIS methods.