Addressing the Log Representativeness Problem using Species Discovery

The analysis of event logs has become a staple in the context of business process management. Insights gained from such an analysis serve to monitor and improve the business processes that generated the logs. Yet, any event log is merely a sample of the past and possible behaviour of a business process, which raises the question of log representativeness: To which extent does the log capture the characteristics of the process that are relevant for the analysis? In this paper, we propose to answer this question using estimators from biodiversity research. Interpreting log representativeness as the completeness regarding distinct properties of a process, we show how to estimate the number of properties often leveraged in process mining in some unknown population. Applying the estimators to real-world event logs, we highlight potential issues in terms of result trustworthiness, also attributing these issues to particular parts of a process.