Deconstructing Diffusion on Tumblr: Structural and Temporal Aspects

Abstract

Online social networks enable collectives of users to create and share content at scale. The diffusion of content through the network, and the resulting information cascades, are phenomena that have been widely investigated on various platforms, which facilitate information diffusion using diverse technical mechanisms, user interfaces and incentives. This paper focuses on Tumblr, an online microblogging social network with a core 'reblogging' functionality that allows information to diffuse across its network by appearing on multiple user blogs. The formation of any cascade network is visible as a list of reblogging events attached as notes to each appearance of the post in the cascade. In this paper, we examine cascade networks on Tumblr, recreated from the series of diffusion events, and analyse them from structural and temporal perspectives. To achieve this, we utilise a cascade construction model that create cascade networks, overcoming problems of a lack of contextual information and missing/degraded data. Finally, we compare cascades in Tumblr with those appearing on other social network platforms. Our analysis shows that popular content on Tumblr creates 'large' cascades that are deep, branching into a large number of separate and long paths, having a consistent number of reblogs at each depth and at each given time.