Modeling the Spread of Influence on the Blogosphere

Abstract

Blogs have become a means by which new ideas and informa- tion spreads rapidly on the web. They often discuss the lat- est trends and echo with reactions to dierent events in the world. The collective wisdom present on the blogosphere is invaluable for market researchers and companies launching new products. In this paper, we evaluate the eectiveness of some of the in uence models on the blogosphere. We also examine the robustness of dierent heuristics in the presence of splogs or spam blogs. Experiments show PageRank based heuristics could be used to select an in uential set of blog- gers such that we could maximize the spread of information on the blogosphere.