Event-Driven Assessment of Currency of Wiki Articles: A Novel Probability-Based Metric
Mathias Klier, Lars Moestue, Andreas Obermeier, Torben Widmann · Journal of the Association for Information Systems · 2021
Wikis provide an easy and efficient way to share and aggregate knowledge. The currency of wikis is of enormous importance since user satisfaction is correlated to the information contained being up-to-date. However, although the automated assessment of currency of wiki articles is an important issue in science and practice alike, existing approaches cannot cope with currency-relevant events as the underlying reasons for outdated articles. To address this issue, we propose a probability-based metric for currency of wiki articles accounting for currency-relevant events. Our event-driven metric reflects the probability that a wiki article is still up-to-date in the sense that no currency-relevant event changed the state of the corresponding real-world entity. We demonstrate the practical applicability and effectiveness of the metric in a real-world setting by analyzing articles from the English Wikipedia. The results show that the provided metric values are reliable and discriminate well between current and outdated articles.