Weighted simplicial complex: A novel approach for predicting small group evolution

The study of small collaborations or teams is an important endeavor both in industry and academia. The social phenomena responsible for formation or evolution of such small groups is quite different from those for dyadic relations like friendship or large size guilds (or communities). In small groups when social actors collaborate for various tasks over time, the actors common across collaborations act as bridges which connect groups into a network of groups. Evolution of groups is affected by this network structure. Building appropriate models for this network is an important problem in the study of group evolution. This work focuses on the problem of group recurrence prediction. In order to overcome the shortcomings of two traditional group network modeling approaches: hypergraph and simplicial complex, we propose a hybrid approach: Weighted Simplicial Complex (WSC). We develop a Hasse diagram based framework to study WSCs and build several predictive models for group recurrence based on this approach. Our results demonstrate the effectiveness of our approach.