This project targets the analysis and the prediction of information quality flaws in Wikipedia articles. The main goals of this project are three-fold: (1) Revealing the nature, the incidence, and the extent of the quality flaws that exist in Wikipedia; (2) Analyzing how quality flaws have evolved since Wikipedia's launch in 2001; and (3) Investigating approaches to automatically predict quality flaws in Wikipedia articles. To achieve these goals we develop technology in the fields of data mining and (semi-supervised) machine learning. The outcome of this project will be relevant for all people who use Wikipedia, including authors, readers, researchers, and data analysts.
Students: Katja Müller, Matthias Busse, Christian Fricke