作者: David Bamman , Noah A. Smith
DOI: 10.18653/V1/D15-1008
关键词:
摘要: Text data has recently been used as evidence in estimating the political ideologies of individuals, including elites and social media users. While inferences about people are often intrinsic quantity interest, we draw inspiration from open information extraction to identify a new task: inferring import propositions like OBAMA IS A SOCIALIST. We present several models that exploit structure exists between assertions they make learn latent positions at same time, evaluate them on novel dataset judged spectrum.