The corpus is composed by 27,152 tokens for training and 4,995 tokens for test. The data are annotated according to the specifications of the semEval 2010 TempEval-2 excercise. The Tempeval2 source data are copyrighted by the various content providers. The annotations are distributed under a Creative Commons Attribution-ShareAlike3.0 Unported License. The data are distributed in two formats: tabbed fields and xml. Detail documentation on the level of annotations is available in the readme file. Official documentation: It-TimeML, TimeML Annotation Guidelines for Italian v. 1.3

You don’t have the permission to edit this resource.