User-connection behaviour analysis in service management using bipartite labelled property graph

Digital transformation is continuously disrupting business models and service delivery. This has resulted in a transition from physical contact in traditional service delivery to digitized user interactions on service platforms. Gleaning insight from patterns of user connections to these services is important for effective service management. First, a user-state bipartite labelled property graph model is constructed for user-connection behaviour analyses. Unlike previous works, the proposed model is neither restricted by the pre-partitioning of data nor the pre-aggregation of edge weights. This model is used to perform flexible versions of static analysis conducted in earlier research works. These ad hoc graph traversal user-connection analyses are typical of real world business scenarios. In this work, they are used to reveal patterns for the range of user interest, user interest intensity and service utilization. Also, an illustration is given of how to perform service recommendation. In addition, an extended model created by enriching a one mode user projection of the bipartite network with connected features is proposed and used to predict user connection behaviour. The proposed approach is effective for the modelling of large user behavioural data sets. The method of analysis is suitable for flexible and expressive real-time analytics in service management. This is particularly useful in domains such data publishing, mobile service usage and telecommunications.