Revision Classification for Current Events in Dutch Wikipedia using a Long Short-term Memory Network
Wikipedia contains pages on unfolding news events, which are updated in near real-time. The revision history of these pages provides a rich source of information to create timelines of the events and to investigate the network of available information through time. However, not every revision is relevant for this purpose. The current paper aims to automatically classify page revisions into various semantic classes correlated to information relevance.
Marijn Schraagen is a computational linguist working on a variety of topics in information retrieval and classification. He obtained a bachelor and master degree in Artificial Intelligence from Utrecht University and a PhD in Data Mining from Leiden University. His application areas of interest include computational historical sociolinguistics and NLP in the legal domain.