A Perspective on the Ubiquity of Interaction Streams in Human Realm

TL;DR

This paper proposes shifting from network-centric modelling of social phenomena to interaction streams, arguing that streaming perspectives align more closely with human cognition and real-time perception. It formalizes social interactions as timestamped tuples $iv^e_{ijk} = (v^e_i, v^e_j, t_k)$ and demonstrates how ego networks, directed weighted edges with $w_{ij} = \frac{n^e_{ij}}{n^e_i}$, and higher-order multi-party interactions can be represented via simplicial complexes or hypergraphs, while also linking to data streams with $S = IS$ and $s_x = iv^e_{ijk}$. The paper outlines two streaming modelling perspectives—per-individual processing and collective-stream modelling without explicit networks—and illustrates the approach with a social learning use case, arguing that streaming methods and preprocessing techniques from data stream mining can improve the realism and efficiency of social process modelling. Although no experiments are provided, the perspective aims to reduce modeling biases from network construction and to offer a framework for integrating cognitive and data-stream insights into social dynamics analysis. This work lays groundwork for comparing streaming versus network-based models and for interdisciplinary exploration of social interactions through stream-oriented representations.

Abstract

Typically, for analysing and modelling social phenomena, networks are a convenient framework that allows for the representation of the interconnectivity of individuals. These networks are often considered transmission structures for processes that happen in society, e.g. diffusion of information, epidemics, and spread of influence. However, constructing a network can be challenging, as one needs to choose its type and parameters accurately. As a result, the outcomes of analysing dynamic processes often heavily depend on whether this step was done correctly. In this work, we advocate that it might be more beneficial to step down from the tedious process of building a network and base it on the level of the interactions instead. By taking this perspective, we can be closer to reality, and from the cognitive perspective, human beings are directly exposed to events, not networks. However, we can also draw a parallel to stream data mining, which brings a valuable apparatus for stream processing. Apart from taking the interaction stream perspective as a typical way in which we should study social phenomena, this work advocates that it is possible to map the concepts embodied in human nature and cognitive processes to the ones that occur in interaction streams. Exploiting this mapping can help reduce the diversity of problems that one can find in data stream processing for machine learning problems. Finally, we demonstrate one of the use cases in which the interaction stream perspective can be applied, namely, the social learning process.

Figures (1)

• Figure 1: An exemplary pipeline used for the link prediction task.