Domain Adaptive Relation Extraction Based on Seeds


 Other Demos
The DARE (Domain Adaptive Relation Extraction) System implements a minimally supervised machine learning framework for extracting relations of various complexity. It
  • starts a bootstrapping from a small set of n-ary relation instances as "seeds"

  • automatically learns pattern rules from parsed data, which can then extract new instances of the n-ary relation and its projections

DARE presents a novel rule representation model which enables the composition of n-ary relation rules on top of the rules for projections of the relation. The compositional approach to rule construction is supported by a bottom-up pattern extraction method.