CounQER: {Coun}ting {Q}uantifiers and {E}ntity-valued P{R}edicates

Predicate constraints of general-purpose knowledge bases (KBs) like Wikidata, DBpedia and Freebase are often limited to subproperty, domain and range constraints. CounQER is a two-step approach, which

  1. identifies set-valued predicates from a given KB predicates via statistical and embedding-based features, and
  2. links counting predicates and enumerating predicates by a combination of co-occurrence, correlation and textual relatedness metrics.

in order to dicover the alignment of counting predicates, like staffSize, and enumerating predicates, like workInstitution-1.

Publications

  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. Uncovering Hidden Semantics of Set Information in Knowledge Bases. Journal of Web Semantics (JWS 2020) (to appear) [pdf]  
  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. CounQER: A System for Discovering and Linking Count Information in Knowledge Bases. (ESWC 2020) [pdf] [demo