Identifying Key Users for Targeted Marketing by Mining Online Social Network

Abstract

The popularity of online shopping highlights the need for targeted marketing. Instead of broadcasting advertisement to an entire online community, targeted marketing aims at key users, namely, influential reviewers whose reviews may affect a large group of his friends, acquaintances or other online customers to buy the product. This paper proposes a method for identifying key users, based on mining of online social networks. We represent social networks as a directed graph of potential customers, which incorporates "web of trust" and "review rating network" on Epinions, and moreover, has a weight associated with each edge to represent the influence of one user on another. We then test a set of algorithms, including general greedy, hill-climbing and centrality-based algorithms, on the real-world social network to identify key users with great influence. We also propose an approximation searching algorithm based on the heuristics information from the above methods. Experimental results showed that if the social network was properly built and associated with sufficient related information, a relatively simple measure was as good as more complex algorithms.