Mingyu Kim, Namgyu Kim, Journal of Internet Computing and Services, Vol. 16, No. 3, pp. 59-70, Jun. 2015
10.7472/jksii.2015.16.3.59, Full Text:  HTML
Keywords: Bigdata Analysis, Site Category Analysis, Site Clustering, Web Log Analysis

Abstract

Many users visit websites every day to perform information retrieval, shopping, and community activities. On the other hand, there is intense competition among sites which attempt to profit from the Internet users. Thus, the owners or marketing officers of each site try to design a variety of marketing strategies including cooperation with other sites. Through such cooperation, a site can share customers' information, mileage points, and hyperlinks with other sites. To create effective cooperation, it is crucial to choose an appropriate partner site that may have many potential customers. Unfortunately, it is exceedingly difficult to identify such an appropriate partner among the vast number of sites. In this paper, therefore, we devise a new methodology for recommending appropriate partner sites to each site. For this purpose, we perform site clustering from the perspective of visitors' similarities, and then identify a group of sites that has a number of common customers. We then analyze the potential for the practical use of the proposed methodology through its application to approximately 140 million actual site browsing histories.

Statistics
Cite this article