mining social networks for viral marketing

Massive quantities of data on very large social networks are now available from blogs, knowledge-sharing sites, collaborative filtering systems, online gaming, social networking sites, newsgroups, chat rooms, etc.
For example, if, in addition to seeing a particular movie myself, I persuade three friends to see it with me, my customer value with respect to that movie has effectively quadrupled, and the movie studio is thus justified in spending more on marketing the movie.
These models allow us to design viral marketing plans that maximize positive word-of-mouth among customers.However, traditional measures of customer value ignore the fact that, in addition to buying products himself, a customer may influence others to buy them.The Network Value of Customers, customer value is usually defined as the expected profit from sales to that customer, over the lifetime of the relationship between the customer and the company.But, while the existence of network effects has been acknowledged in the marketing literature, they have generally been considered to be unquantifiable, particularly at the level of individual customers.We have begun to build social network models at this scale, using data from the Epinions knowledge-sharing site, the EachMovie collaborative filtering system, and others 1,.These networks typically number in the tens of thousands to millions of nodes, and often contain substantial quantities of information at the level of individual nodes, sufficient to build models of those individuals.We call the network value of a customer the expected increase in sales to others that results from marketing to that customer.Clearly, ignoring the network value of customers, as is done in traditional direct marketing, may lead to very suboptimal marketing decisions.
In the past, this was largely due to lack of data: the networks available for experimental study were small and few, gagner telephone portable and contained only minimal information about each node.
Customer value is of critical interest to companies, because it determines how much it is worth spending to acquire a particular customer.
Assembling these models into models of the larger network they are part of gives us an unprecedented level of detail in social network analysis, with the corresponding potential for new understanding, useful predictions, and their productive use in decision-making.In our experiments, this makes it possible to achieve much higher profits than if we ig- nore interactions among customers and the corresponding network effects, as traditional marketing does.This is what is changed by the data sources now available.Our models enable us to measure the network value of a customer.Conversely, if I tend to make decisions on what movies to see purely based on what my friends tell me, marketing to me may be a waste of resources, which would be better spent marketing to my friends.Traditionally, social network models have been descriptive, rather than predictive: they are built at a very coarse level, typically with only a few global parameters, and are not useful for making actual predictions of the future behavior of the network.

Fortunately, the rise of the Internet has changed this dramatically.
One of the major applications of data mining is in helping companies determine which potential customers to market.
If the expected profit from.