• Korean Business Review
• /
• v.20 no.1
• /
• pp.145-168
• /
• 2007

Recommender system helps customers to find easily items and helps the e-biz companies to set easily their target customer by automated recommending process. Recommender systems are being adopted by several e-biz companies and from these systems, both of customers and companies take some benefits. This study sets several thresholds to the similarity weight, which indicates a degree of similarity of two customers' preference, to improve the performance of prediction accuracy. According to the threshold, the accuracy of prediction is being improved but some threshold setting shows the reduction of the prediction rate, which is the coverage. This coverage reduction has male effect on the prediction accuracy of customers, so more study on the prediction accuracy of recommender system and to maximize the coverage are needed.

• Journal of the Korea Society of Computer and Information
• /
• v.8 no.1
• /
• pp.140-148
• /
• 2003

It is expected that the role and market share of B2B Electronic Commerce between countries increase in the future. To motivate this new type of marketing, civil-leading flexible regulation frame is needed than any other factors including its legislation and administration for activating each companys cooperation mind and friendly relationship also. Government policies should focus on coordination of competing circumstances and participating to international standardization working scope In addition, domestic situation of B2B Electronic Commerce and political perspectives are to be analyzed and proposed that contribute to broaden market share. These aims are obtained by scrutinizing some critical factors of current B2B market: companys trading characteristics, possibility of introducing to trading scope, and B2B Electronic Commerce Policy etc.

• Journal of the Korea Society for Simulation
• /
• v.15 no.3
• /
• pp.23-30
• /
• 2006

A recommender system for E-commerce site receives information from customers about which products they are interested in, and recommends products that are likely to fit their needs. In this paper, we investigate several methods for large-scale product purchase data for the purpose of producing useful recommendations to customers. We apply the traditional data mining techniques of cluster analysis and collaborative filtering(CF), and CF with reduction of product-dimensionality by use of latent semantic indexing(LSI). If reduced product-dimensionality obtained from LSI shows a similar latent trend of customers for buying products to that based on original customer-product purchase data, we expect less computational effort for obtaining the nearest-neighbor for target customer may improve the efficiency of recommendation performance. From simulation experiments on synthetic customer-product purchase data, CF-based method with reduction of product-dimensionality presents a better performance than the traditional CF methods with respect to the recall, precision and F1 measure. In general, the recommendation quality increases as the size of the neighborhood increases. However, our simulation results shows that, after a certain point, the improvement gain diminish. Also we find, as a number of products of recommendation increases, the precision becomes worse, but the improvement gain of recall is relatively small after a certain point. We consider these informations may be useful in applying recommender system.

• Proceedings of the CALSEC Conference
• /
• 1999.07a
• /
• pp.233-242
• /
• 1999

갈수록 치열해져 가는 글로벌 경쟁환경에서 우리나라 기업들이 전자상거래를 통해 세계적 전자/정보통신 업체들과 경쟁할 수 있는 환경을 민간업계와 정부가 협력하여 조속히 구축, 운영해야 함.(중략)

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.4
• /
• pp.803-813
• /
• 2013

The purchase of items in e-commerce is a little bit different from that of items in off-line. The recommendation of items in off-line is conducted by salespersons' recommendation, However, the item recommendation in e-commerce cannot be recommended by salespersons, and so different types of methods can be recommended in e-commerce. Recommender system is a method which recommends items in e-commerce. Preferences of customers who want to purchase new items can be predicted by the preferences of customers purchasing existing items. In the recommender system, the items with estimated high preferences can be recommended to customers. The algorithm of collaborative filtering is used in recommender system of e-commerce, and the list of recommended items is made by estimated values, and then the list is recommended to customers. The dataset used in this research are 100k dataset and 1 million dataset in Movielens dataset. Similar results in two dataset are deducted for generalization. To suggest a new algorithm, distribution features of estimated values are analyzed by the existing algorithm and transformed algorithm. In addition, respondent'distribution features are analyzed respectively. To improve the collaborative filtering algorithm in neighborhood recommender system, a new algorithm method is suggested on the basis of existing algorithm and transformed algorithm.

• Proceedings of the CALSEC Conference
• /
• 1999.07a
• /
• pp.373-382
• /
• 1999

EC(Electronic Commerce) : 상업적인 데이터 교환 TOOL ▶ 기업과 기업간 또는 기업과 소비자간에 통합적인 자동화된 정보체계 환경하에서 판매, 생산, 구매, 재무, 운송, 행정, ▶ 서비스등의 상거래를 전자적으로 실현하는 행위 ▶ B to C (Business to Consumer) : Shopping Mall ▶ B to B (Business to Business) : EDI(중략)

• 한국IT서비스학회:학술대회논문집
• /
• 2007.11a
• /
• pp.616-620
• /
• 2007

전자상거래에서 사용되고 있는 추천시스템은 사용자들의 프로파일과 이들의 정보를 바탕으로 사용자가 선호할 만한 아이템을 추천한다. 추천시스템에서 널리 사용되고 있는 협력적 필터링 방식은 사용자들 사이의 선호도 평가치를 비교하여 유사 사용자를 선택하고, 아이템에 대한 유사 사용자의 선호도 평가치를 기반으로 하여 추천하고자 하는 아이템에 대한 사용자의 선호도를 예측하는 것이다. 하지만 사용자의 선호도가 적은 데이터로 인한 희소성 문제는 추천시스템의 성능을 저해하는 요인으로 작용하고 있다. 이러한 희소성의 문제는 선호도 평가 자료에 나타난 아이템들의 총수에 비하여 사용자가 선호한 아이템의 수가 아주 적기 때문에 발생하며, 새로운 사용자의 경우에는 아이템에 대한 선호도 평가치가 없어 유사 사용자를 선택할 수가 없어 나타나며 심한 경우에는 아이템을 전혀 추천할 수 없게 된다. 이리할 추천 시스템의 희소성문제를 해결차기 위한 방법은 희소성이 높은 데이터들에 대한 희소성을 감소시키는 것이다. 따라서 본 논문에서는 아이템에 대한 희소성을 조사하여 협력적 필터링에서 희소성 아이템이 MAE에 미치는 영향을 분석하였다. 그리고 희소성 문제를 완화하여 예측 정확도를 높이기 위한 방법으로 선호도가 적은 아이템에 대해 희소성을 최소화하는 연구와 이에 따라 희소성과 MAE의 값을 개선하는 방법을 제안한다.

• Journal of Korea Multimedia Society
• /
• v.8 no.1
• /
• pp.128-136
• /
• 2005

In this paper, we designed and implemented the general-purpose document repository manager for the ebXML-based e-business framework that support the rapid translation of various types of documents and provide the flexibility corresponding to the framework changes. Also, we automated the manually-processed e-business document translation process by designing the adaptor module for the interoperability of legacy systems. In the regards, our designed general purpose document manager makes it possible to integrate the existing e-commerce systems that are operated on the heterogeneous frameworks, and can be used on global e-business solution.

• The Journal of Society for e-Business Studies
• /
• v.14 no.3
• /
• pp.131-151
• /
• 2009

Natural interface is important to select and provide the services in ubiquitous smart space such as u-plant, u-distribution. Augmented Reality(AR) has recently begun to receive attention as a realization tool for natural interface. AR provides virtual object on real environment and it differs from virtual reality. When AR is used, it has advantage to provide information intuitively and collaboratively. However AR is rarely used in e-commerce domain of ubiquitous smart space, and it has limitation which predefined information and services provide in a static manner. Hence, the purpose of this paper is to propose a methodology of AR based e-commerce which provides personalized association service by considering user's dynamic context. To do so, association algorithm is developed based on Rescorla-Wagner model and Galton's free association test.

• 한국IT서비스학회:학술대회논문집
• /
• 2008.05a
• /
• pp.516-520
• /
• 2008

협력적 필터링을 이용한 추천시스템은 인터넷 기반 전자상거래에서 좋은 추천 도구로 사용되고 있다. 협력적 필터링 방식은 고객의 선호도를 조사하여 이를 바탕으로 이웃 고객을 선정하고 이들에 대한 선호도를 수집하여 고객이 좋아할 만한 상품을 추천하는 기법이다. 이웃 고객에 대한 정보를 이용하여 추천에 사용하므로 이웃고객이 적은 경우 추천시스템의 예측에 어려움이 생긴다. 본 논문은 추천시스템의 예측 정확도를 높이기 위한 방법으로 희소성이 있는 상품을 우선 선정하고 그들 상품에 대한 선호도를 조사하였다. 그리고 이들에 대한 선호를 나타낸 고객들을 선별하여 추천시스템의 예측 정확도를 향상시키는 방법을 제안한다.