AI Seminar

Graduate Student Presentations

AI Graduate Students

The internet provides different media for the diffusion of information. One such medium are weblogs that link together in a complex structure through which new ideas and discourse can flow. We can infer and visualize the paths specific pieces of information take through the network and also distinguish how different kinds of information propagate. A second medium is email, which allows, for example, customers to recommend products to one another. We present an analysis of a person-to-person recommendation network, consisting of 4 million people who made 16 million recommendations on half a million products. We observe the propagation of recommendations and the cascade sizes, which can be explained by a stochastic model. We find that the structure of the network influences the success of recommendations. For example, people recommending to many others will influence more purchases, but there is a saturation effect. Furthermore, while on average recommendations are not very effective at inducing purchases and do not spread very far, there are product and pricing categories for which viral marketing seems to be very effective.

Sponsored by

AI Laboratory