Learning Rules for Textual Entailment

Опубликовано 6 сентября 2016, 17:07

The design of models that learn textual entailment recognizers from annotated examples is not simple as it requires the modeling of semantics involved in the interaction of pairs of text fragments. In this talk, we firstly introduce the class of pair feature spaces which allow supervised machine learning algorithms to derive first-order rewrite rules from annotated examples. In particular, we propose the syntactic and the shallow semantic pair feature spaces.