Unsupervised Entailment Detection between Dependency Graph Fragments

dc.creatorRei, M
dc.creatorBriscoe, T
dc.date2018-03-22T16:20:45Z
dc.date2018-03-22T16:20:45Z
dc.date2011-06-23
dc.date.accessioned2026-08-03T03:07:56Z
dc.descriptionEntailment detection systems are generally designed to work either on single words, relations or full sentences. We propose a new task – detecting entailment between dependency graph fragments of any type – which relaxes these restrictions and leads to much wider entailment discovery. An unsupervised framework is described that uses intrinsic similarity, multi-level extrinsic similarity and the detection of negation and hedged language to assign a confidence score to entailment relations between two fragments. The final system achieves 84.1% average precision on a data set of entailment examples from the biomedical domain.
dc.formatapplication/pdf
dc.identifierhttps://www.repository.cam.ac.uk/handle/1810/274250
dc.identifier10.17863/CAM.21358
dc.identifier.urihttps://repo.dare.co.zw/handle/123456789/173143
dc.languageeng
dc.titleUnsupervised Entailment Detection between Dependency Graph Fragments
dc.typeConference Object

Files