User’s Behavior in Digital Libraries: Process Mining Exploration

Abstract

In digital libraries, users have diverse ways to perform their research according to their objectives. In order to have access to target documents, users almost make sequences of activities that can be considered as processes. It would be useful to represent the behavior of users in their information seeking processes in order to have a global view on users’ interactions with digital libraries and to effectively answer to their requirements. Discovered processes in digital libraries are almost self-defined that may vary in term of significance, structure and results. It is therefore important to automatically build a model that describe users’ behaviors. In this paper, we tested six state-of-the-art process discovery techniques on two datasets collected from users’ interactions with a cultural heritage digital library. Our preliminary evaluation presents the best performing model that can be used to optimize the exploration of documents in digital libraries.