Titre: A natural language interface for querying process execution data

Abstract: In this poster, we showcase our research on a natural language interface that facilitates the querying of process execution data for end users. Specifically, we focus on assisting business users, who may not possess the technical expertise required to efficiently query process execution data using traditional methods. We start first by presenting the graph metamodel that we designed for the storage of process data based on a Labeled Property Graph. We then present our hybrid pipeline that combines machine learning and rule-based techniques to automatically generate a Cypher query from a user’s question, which can then be executed over the stored graph to yield the desired results.