Decision Boundaries in User Journey Event Logs

Abstract

Service providers may systematically record and analyze how users interact with a service. User journeys model this interaction from the user’s perspective. We can think of user journeys in terms of multi-party event logs in which events are triggered by two independent parties, the user and the service provider, both controlling their share of actions. Multi-party event logs enable the automated construction of weighted two-player games, which can be used to analyze the user journeys. However, the games constructed for complex user journeys may get large, which makes visual understanding challenging. To reduce the size of these games, we develop an analysis technique for a game’s decision boundary, at which the outcome of the game is determined. Decision boundaries identify subgames that are equivalent to the full game with respect to the final outcome of user journeys. The paper formalizes decision boundaries for user journey games with history refinements, and studies properties of decision boundaries that simplify their analysis. The decision boundary analysis from multi-party event logs has been implemented in a modular tool, which can be connected to existing process mining pipelines, and evaluated on benchmark event logs revealing changes in the user journey and the interaction between users and service provider.