By Topic

Customer-Driven Content Recommendation Over a Network of Customers

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Hyea Kyeong Kim ; School of Management, Kyung Hee University, Seoul, Korea ; Young U. Ryu ; Yoonho Cho ; Jae Kyeong Kim

As the Web evolves into an ecological platform of information, people, and technologies, its usage paradigm has gradually shifted so that the importance of its participative role is observed. Users contribute by uploading multimedia content, writing wiki pages, and posting blog articles. As the effect of user participation (the word-of-mouth effect) in the Internet becomes a factor influencing firms' success, firms search for ways to utilize blogs, social networks, and other Internet resources. To actively make use of the online word-of-mouth effect, firms must structure preference-based customer networks so that local interaction happens among closely related customers and effective propagation of ideas or diffusion of products can be achieved. In this paper, we propose a recommendation technique utilizing the fast diffusion and information sharing capability of a large customer network. The proposed method [described as the customer-driven recommender system (CRS)] follows the collaborative filtering (CF) principle but performs distributed and local searches for similar neighbors over a customer network in order to generate a recommendation list. In order to validate the effectiveness and efficiency of the proposed method, we build customer networks for the recommendation of digital content and tangible products from two real data sets and compare the proposed method against the traditional system based on CF. Experimental results show that the local search mechanism of the CRS is computationally more efficient than but equally as accurate as the global search mechanism of the traditional recommender system.

Published in:

IEEE Transactions on Systems, Man, and Cybernetics - Part A: Systems and Humans  (Volume:42 ,  Issue: 1 )