Real-time Population of Knowledge Bases: Opportunities and Challenges

Dynamic content is a frequently accessed part of the Web. However, most information extraction approaches are batch-oriented, thus not effective for gathering rapidly changing data. This paper proposes a model for fact extraction in real-time. Our model addresses the difficult challenges that timely fact extraction on frequently updated data entails. We point out a naive solution to the main research question and justify the choices we make in the model we propose.