UW Event Factuality Dataset
This dataset contains annotations of text from the TempEval-3 corpus. The annotation is described in:
In Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2015
The dataset is in the BRAT standoff format (http://brat.nlplab.org/standoff.html). It is split into train, development, and test sets. A visualization of the data can be found at http://appositive.cs.washington.edu/brat/index.xhtml#/fact
If you are looking for the code to replicate the experiment, see https://bitbucket.org/kentonl/factuality-experiments.
For questions and inquiries, please contact Kenton Lee.