Editing Behavior Analysis for Predicting Active and Inactive Users in Wikipedia
These days, user-generated content platforms such as social media, question-answering Websites, and open collaboration systems are a source of information for many. These platforms survive, thanks to the pool of active contributors who generate content. As a consequence, they continuously face the problem of acquiring new users and retain them in the platform.
In this paper, we study the case of English Wikipedia, a well-established open collaboration system, and study the problem of predicting whether or not an editor will become inactive and stop contributing to the encyclopedia. Knowing this information can help the administrative community to perform engaging actions on time to keep users contributing longer.
We propose a predictive model leveraging contributors’ editing behavior to identify active vs. inactive Wikipedia users. Our experiments show that our method achieves an AUROC of at least 0.97 in predicting editors who will become inactive and can predict inactive users earlier than the baseline.
Arelli, Harish; Spezzano, Francesca; and Shrestha, Anu. (2019). "Editing Behavior Analysis for Predicting Active and Inactive Users in Wikipedia". Influence and Behavior Analysis in Social Networks and Social Media, 127-147. https://dx.doi.org/10.1007/978-3-030-02592-2_7