MSc Thesis Defence: Owen Fairholm

The School of Computer Science is pleased to announce the MSc Defence of Owen Fairholm, to be presented on December 18, 2014, at 9:30 in Reynolds 219. The presentation is titled "Annotation of Temporal Relations Using Markov Logic Networks and Temporal Centering".

Title

Annotation of Temporal Relations Using Markov Logic Networks and Temporal Centering

Abstract

In recent years annotating temporal relations has been a focus by many people in the area of Temporal Information Processing. By labeling these relations within text, the events and times within can be ordered. This leads to event occurrence being established, and event order settled. Methods such as Markov Logic Networks and other have achieved very good results when presented with this task. We purpose a combined approach of Markov Logic Networks with suggestions provided by a Temporal Centering technique. These will combine to show the benefits of both the probabilistic nature of Markov Logic Networks and the context based knowledge of Temporal Centering. By combining these two approaches we allow the Markov Logic Network to achieve a better understanding of the text, and therefore improve the results of its labeling. The results obtained from using these two approaches are competitive with the state-of-the-art systems submitted to the TempEval commissions.

Advisors: Fei Song