Overview of CLEF 2019 Lab ProtestNews: Extracting Protests from News in a Cross-context Setting

We present an overview of the CLEF-2019 Lab ProtestNews on Extracting\nProtests from News in the context of generalizable natural language processing.\nThe lab consists of document, sentence, and token level information\nclassification and extraction tasks that were referred as task 1, task 2, and\ntask 3 respectively in the scope of this lab. The tasks required the\nparticipants to identify protest relevant information from English local news\nat one or more aforementioned levels in a cross-context setting, which is\ncross-country in the scope of this lab. The training and development data were\ncollected from India and test data was collected from India and China. The lab\nattracted 58 teams to participate in the lab. 12 and 9 of these teams submitted\nresults and working notes respectively. We have observed neural networks yield\nthe best results and the performance drops significantly for majority of the\nsubmissions in the cross-country setting, which is China.\n