How Did You Get to Know That? A Traceable Word-of-Mouth Algorithm

Word-of-mouth communication has been shown to play a key role in a variety of environments such as viral marketing and virus spreading. A family of algorithms, generally known as information spreading algorithms or word-of-mouth algorithms, has been developed to characterize such behavior. However, they have limitations, including the inability to: (1) capture when the communications or contacts take place and (2) explain where the influence comes from. These drawbacks have limited the studies about how the spreading of influence takes place in social networks. In this paper, we present a new word-of-mouth algorithm that considers the temporality of the communications and keeps track of how influence travels over the social network. We validate the proposed algorithm via simulations of word-of-mouth traces on call detailed records, in order to model how influence spreads. Our results indicate that (1) static factors of social networks are not enough to model influence and (2) there seems to be statistical invariants of how influence spreads in a network.