Web3 Social Platforms: Modeling, Mining and Evolution

Web3, one of the arising paradigms which may rule the future Web, is also representing a source of big data stored in the underlying blockchains. Many different research fields are benefiting from these large collections of temporal and heterogeneous data, which capture different aspects of the interactions among people and between people and Web3 platforms. Specifically, since each piece of information is validated and timestamped, Web3 platforms are becoming an invaluable source for understanding the dynamics of these techno-social systems at a high temporal resolution. In this contribution, we focused on the analysis of the evolution of the networked structure of Web3 social networks through the lens of discrete choice models, and on the changes in the structure of the relationships after a shocking event has occurred on the platform - namely a hard-fork in the supporting blockchain. To support large-scale analysis, we represent Web3 platform data as temporal multigraphs manageable by modern graph database management systems. The main findings, which represent a summary of our effort in mining data from Web3 platforms, highlight some interesting aspects: i) when applied to Web3 social networks, discrete choice models allow us to decompose the evolution of social networks into different growing mechanisms, which are quite stable during the observation period; and ii) in a stratified context, such as Web3 platforms, interactions resulting from economic actions, such as transfers or loans of crypto-tokens, are as important as social relationships to predict how users will behave during a shocking event. These are a few examples of how Web3 social platforms may represent a challenging playground for a more in-depth understanding of the users' behaviors when social and economic interactions are strictly intertwined.