Posting contents online is at everybody's fingertips. As a result any person can easily distribute information to influence in the people's perception. Misinformation have become trendy as it influences politics and economy all around the globe.
Our efforts on fighting online misinformation include two fronts. In front CheckThat! we are promoting the research on fact checking by creating evaluation frameworks for researchers to design models which grab evidence from the Web and verify the factuality of a claim. In front proppy, we aim at making propaganda in online news evident to the general audience. We have developed supervised models to identify whether a news article contains high levels of propaganda. Nowadays, we focus on the development of models to identify specific propagandist fragments.
I will offer details on the inception of proppy and CheckThat! and on our ongoing contributions to limiting the impact of misinformation through these two fronts.
Alberto Barron-Cedeno obtained his PhD from Universitat Politecica de Valencia (Spain). He is senior assistant professor at Universitat di Bologna (Italy). Previously, we was a Scientist at Qatar Computing Research Institute (Qatar) and the Alain Bensoussan fellow at Universitat Politecnica de Catalunya (Spain). He is interested in the automatic analysis of diverse qualities of texts, such as originality, relevance, and intent; also across languages. Alberto has published 60+ papers in the most important forums of NLP and IR (e.g., ACL, EMNLP, SIGIR, ECIR, IP&M, LRE, KNOSYS). He has co-organized various editions of the labs PAN and CheckThat! at CLEF and is currently organising a shared task at SemEval on propaganda identification.