• 한국컴퓨터정보학회논문지
• /
• 제23권12호
• /
• pp.219-226
• /
• 2018

Collaborative filtering has been most popular approach to recommend items in online recommender systems. However, collaborative filtering is known to suffer from data sparsity problem. As a simple way to overcome this problem in literature, Jaccard index has been adopted to combine with the existing similarity measures. We analyze performance of such combination in various data environments. We also find optimal weights of factors in the combination using a genetic algorithm to formulate a similarity measure. Furthermore, optimal weights are searched for each user independently, in order to reflect each user's different rating behavior. Performance of the resulting personalized similarity measure is examined using two datasets with different data characteristics. It presents overall superiority to previous measures in terms of recommendation and prediction qualities regardless of the characteristics of the data environment.

• 한국멀티미디어학회논문지
• /
• 제13권10호
• /
• pp.1453-1462
• /
• 2010

본 연구는 IPTV 환경에서 사용자의 취향에 맞는 VOD 프로그램을 추천할 수 있는 시스템을 새로이 제안하였다. 추천 알고리즘으로는 협업 필터링 기법을 사용하였다. 사용자의 프로그램 선호 취향을 나타내는 사용자 프로파일(user profile)은 사용자와 유사한 취향의 이웃 사용자들의 프로그램 선호도와 중분류 선호도 그리고 사용자 유사도를 감안하여 1주 단위로 갱신하였다. 제안 시스템의 성능평가를 위해 시청률 조사기관인 닐슨 리서치의 24주분 지상파 및 케이블 방송 시청 데이터를 IPTV 형식에 맞게 재구성하여 사용하였으며, 다양한 실험을 통해 그 실용성을 입증하였다. 실험결과 사용자 유사도 가중치를 사용하며, 그룹 크기가 5명 그리고 추천 프로그램 수가 5개 일 때 최적의 성능을 나타내었다.

• 정보처리학회논문지:컴퓨터 및 통신 시스템
• /
• 제2권11호
• /
• pp.475-482
• /
• 2013

추천 시스템들에 대한 여러 방법들이 연구되고 있다. 개인화와 추천 시스템 중에서 가장 성공적인 방법은 협업 필터링이다. 협업 필터링은 고객들의 프로파일 정보를 기반으로 추천을 하므로 데이터가 충분하지 않다면 항목을 추천하는데 있어서 희박성의 문제가 제기된다. 본 연구에서는 희박성의 문제를 해결하는 방법으로 가중치를 가진 개인 성향을 협업 필터링에 활용하는 방법을 제안한다. 본 연구에서 가중치를 가진 최적의 개인 성향을 찾기 위해 공개 데이터인 MovieLens Data를 이용하여 성능 평가하였다. 실험 결과 본 연구에서 제안한 가중치를 가진 개인 성향들로 구축된 시스템이 기존의 개인 성향들을 이용한 시스템보다 향상된 성능을 보였다.

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2007년도 한국지능정보시스템학회
• /
• pp.179-188
• /
• 2007

디지털 기기 가 보편 화 되 면서 많 은 디지털 컨텐츠가 생성되고 있다. 또한, 인터넷 서비스의 발전으로 이들 컨텐츠를 과거에 비해 손쉽게 웹 상에 개제할 수 있게 되 었다. 따라서, 많은 컨텐츠를 추 천해 주기 위해 추천 시스템에 관한 연구가 활발히 진행되고 있다. 이들 컨텐츠가 기존의 텍스트 기반에서 사진이나 동영상, 사운드 등 컴퓨터가 자동으로 내용을 파악하기 힘든 컨텐츠로 변화하면서, 내용의 파악이 필요 없 는 협력적 여 과(Collaborative Filtering)가 추천 시스템에서 유 용하게 이 용될 수 있다. 또한 web 2.0의 영향으로 컨텐츠를 분류하고 재검색을 용이하게 하기 위해 태깅(tagging)을 제공하는 서비스가 많아지고 있다. 본 논문에서는 내용 파 악이 힘든 컨텐츠의 효과적인 추천을 위해 협력적 여과(Collaborative Filtering)와 협력적 태깅(Collaborative Tagging)을 접목시킨 방법을 제안하고, 전통적인 협력적 여과 방법과 제안한 방법의 비교 실험을 통하여 협력적 여과 방법에서의 태 깅의 효과에 대 해 논한다.

• 한국정보처리학회논문지
• /
• 제6권11S호
• /
• pp.3420-3433
• /
• 1999

Since most of CSCW(Computer-Supported Cooperative Work) systems have been developed for particular computing platforms, they are usable only within the specific organizations supporting those particular platforms. Recently, according to the rapid growth and continuing success of the World-Wide Web(WWW or Web) which offers a globally accessible platform-independent infrastructure, many CSCW systems has been constructed and is being developed on the basis of the Web. As one of such CSCW systems, in this paper, we describe the design and implementation of the CoSpace system. The CoSpace system provides shared workspaces, which enable members of a work group easily share information for collaborative work through their Web Browsers. The shared workspaces support information sharing and management between users, and also support event monitoring and synchronization between collaborative works.

• Journal of Astronomy and Space Sciences
• /
• 제19권2호
• /
• pp.107-122
• /
• 2002

Multi-body spacecraft attitude and configuration control formulations based on the use of collaborative control theory are considered. The control formulations are based on two-player, nonzero-sum, differential game theory applied using a Nash strategy. It is desired that the control laws allow different components of the multi-body system to perform different tasks. For example, it may be desired that one body points toward a fixed star while another body in the system slews to track another satellite. Although similar to the linear quadratic regulator formulation, the collaborative control formulation contains a number of additional design parameters because the problem is formulated as two control problems coupled together. The use of the freedom of the partitioning of the total problem into two coupled control problems and the selection of the elements of the cross-coupling matrices are specific problems ad-dressed in this paper. Examples are used to show that significant improvement in performance, as measured by realistic criteria, of collaborative control over conventional linear quadratic regulator control can be achieved by using proposed design guidelines.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 2006년도 춘계학술대회 논문집
• /
• pp.619-620
• /
• 2006

In this paper, the CAE Service System for Collaborative Engineering Environment with web services and Multi-frontal Method has been investigated and developed. The enabling technologies such as SOAP and .NET Framework play great roles in the development of integrated distributed application software. In addition to the distribution of analysis modules, numerical solution process itself is again divided into parallel processes using Multi-frontal Method for computational efficiency. We believe that the proposed approach for the analysis can be extended to the entire product development process for sharing and utilizing common product data in the distributed engineering environment.

• 대한안전경영과학회:학술대회논문집
• /
• 대한안전경영과학회 2005년도 춘계학술대회
• /
• pp.81-89
• /
• 2005

The purpose of this study is to develop S/W of the web-based collaborative warehouse management system using RFID(Radio Frequency Identification). This S/W System supports the realtime inventory management and collaborative operations of relational companies in SCM. We look for benefits-reduce inventory levels, maxmize use of warehouse space, decrease the logistic cost, and increase competitive power of company, using this S/W

• 한국지능정보시스템학회:학술대회논문집
• /
• 한국지능정보시스템학회 2003년도 춘계학술대회
• /
• pp.416-425
• /
• 2003

Recommender systems are a personalized information filtering technology to help customers find the products they would like to purchase. Collaborative filtering (CF) has been known to be the most successful recommendation technology. However its widespread use in e-commerce has exposed two research issues, sparsity and scalability. In this paper, we propose several hybrid recommender procedures based on web usage mining, clustering techniques and collaborative filtering to address these issues. Experimental evaluation of suggested procedures on real e-commerce data shows interesting relation between characteristics of procedures and diverse situations.

• 한국멀티미디어학회논문지
• /
• 제22권8호
• /
• pp.897-912
• /
• 2019

The growth of domestic fashion industry market went sluggish affected by long-term slow growth and shrinking consumer sentiment since 2012, and the entire market has shown a slight growth since 2017. Yet only some conglomerates or global businesses experience this growth. Small and medium-sized fashion companies or rising fashion designers in South Korea, face real limitations and problems poor capital, lack of professional workforce, experience, information and educational opportunity as well as difficulty of acquiring distribution network, which lead to further polarization between conglomerates and small businesses. Thus, this study proposed mobile virtual viewer system-based collaborative application prototype capable of overcoming limitations and problems drawn by analyzing business environment of domestic brands launched by rising fashion designers. Especially, through applied template-based 3D virtual fashion design technology and performing the entire process in the virtual reality beyond spatial and temporal restraints, it is assumed that more effective and efficient outcomes can be obtained compared to the previous method.