Corpus Gold Standard TAGFACT v.0.1
收藏DataCite Commons2025-06-10 更新2025-04-09 收录
下载链接：
https://dataverse.csuc.cat/citation?persistentId=doi:10.34810/data619
下载链接
链接失效反馈官方服务：
资源简介：
The Corpus Gold Standard TAGFACT v.0.1 is a Spanish corpus manually annotated with factuality information. The corpus consists of 11 news items (basically political), with a total of 5,240 words (approximately 160 sentences) extracted from newspapers. They are all newspapers published in Spain and, predictably, are mostly in peninsular Spanish.
The corpus has been manually annotated with information about factuality of 633 predicates. Six linguists worked on it, all of whom had undergone training. The analysis of inter-annotator agreement was made in pairs, a total of 30, and, on average, the result was acceptable.
Two XML versions of the Corpus Gold Standard TAGFACT v.0.1 are available. In each one different tagsets (defined with more or less abstraction) have been used, but equivalences have been stablished.
提供机构：
CORA.Repositori de Dades de Recerca
创建时间：
2023-02-08
