Fork-based user migration in Blockchain Online Social Media

Nowadays, Online Social Media (OSM) are among the most popular web services. Traditional OSM are known to be affected by serious issues including misinformation, fake news, censorship, and privacy violations, to the point that a pressing demand for new paradigms is raised by users all over the world. Among such paradigms, the concepts around the Web 3.0 are fueling a new revolution of online sociality, pushing towards the adoption of innovative and groundbreaking technologies. In particular, the decentralization of social services through the blockchain technology is representing the most valid alternative to current OSM, enabling the development of rewarding strategies for value redistribution, and fake news detection. However, the so-called Blockchain Online Social Media (BOSMs) are far from being mature, with different platforms that continually try to redefine their services in order to attract larger audiences, thus causing blockchain forks and massive user migrations, with the latter dominating the dynamics of the current OSM landscape, too. In this paper, we deal with the evolution of BOSMs from the perspective of user migration across platforms as a consequence of a fork event. We propose a general user migration model applicable to BOSMs to represent the evolution patterns of fork-based migrations, the multi-interaction structural complexity of BOSMs, and their growth characteristics. Within this framework, we also cope with the task of predicting how users will behave in the case of a fork, i.e. they will remain on the original blockchain or they will migrate to the new one. We apply our framework to the case study of the Steem-Hive fork event, and show the importance of considering both social and economic information, regardless of the learning algorithm considered. To the best of our knowledge, this is the first study on blockchain fork and its related user migration.