Information Diffusion and Provenance in Online Social Media Networks

Traditional data mining techniques are going through an extensive research so as to suit their application in the dynamically evolving social media networks which are not only large in size but also require real time processing of data streams. One such application which is of social, industrial and political interest is to maximize (sometimes even minimize) information diffusion in online social media networks. Word-of-mouth (w-o-m) communications have been largely employed in a wide range of application scenarios ranging from marketing strategies to creating mass awareness. In recent years, we saw an explosion in the growth of online social networks with many successful service providers like Facebook, twitter, YouTube, LinkedIn etc. ruling the market. The use of online social networks is not limited to fun and building professional networks. Online social networks are now seen as a powerful medium to effect the social and political environment of a country and the world as a whole. Information on these social networking sites sometimes needs to be propagated very effectively and quickly to fulfill the concerned strategies. It is equally important that given some information present on social network, we are able to derive its origin so that rumours can be distinguished from truth. In this paper, we discuss some of the issues and challenges in information diffusion and provenance from online social networks point of view. We also discuss some of the approaches and models that have been used for information diffusion and provenance in online social networks in recent past. Some of these approaches have been evaluated by researchers for specific social network services but we will discuss them in a general sense.