Influential user mining for viral and target marketing in social network through PageRank and Bayesian inference

Abstract

Social networks have become one of the most important focuses for almost all Business strategies due to massive increase of potential sales using Viral marketing. The chief role played in these networks are the influential users, the actual market movers in any critical networks. Finding these users demands suitable approaches to take that oftentimes depends on the criteria of a social network along with the study of user behavior. Target market can be referred to as a community of people who are most likely to purchase some specific products and/or who have the highest odds of spreading the product. They are most likely to buy the product, somehow be in need of it or have a high record of being motivated by their idols, i.e. who they follow. They tend to have some common demo-graphical and behavioral characteristics (in that network) and thus the focus lies on what characteristics they share in that network which the business is interested in. Viral marketing is popular nowadays as it has its own business value. It can be termed as a strategy to find how customers spread messages about the product with other people in their social network, like the same way a virus spreads from one person to another. In this research proposal, we focus on target or viral marketing by studying efficient influential user mining procedures in twitter networks. We propose the famous PageRank algorithm and Bayesian Inference to find the best influential users in the network.