D5
Databases and Information Systems
Illustrative example of count information

Count Information

Cont information is the relation between an entity and a set of entities that can be expressed individually, as X is an employee of Z, Y is an employee of Z or as a cardinality Z has N employees. Identifying such information in text and knowledge bases (KBs) and linking the two variants can

  1. identify incomplete entity enumerations
  2. ground counts in representative entities
  3. identify KB inconsistencies
  4. enhance downstream tasks like search and question answering

Most research focus on numerical facts without joining the dots between numerical facts specifying cardinal information and their corresponding entity-to-entity grounding facts. Popular general-purpose KBs have limited ability to recognize such predicates. 

Publications

  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. CoQEx: Entity Counts Explained. (WSDM 2023) [pdf] [poster] [video] 
  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. Answering Count Questions with Structured Answers from Text. (JoWS 2022) [arxiv|journal]
  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. Answering Count Queries with Explanations. (SIGIR 2022) [pdf] [poster] [code|data]
  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. Uncovering Hidden Semantics of Set Information in Knowledge Bases. Journal of Web Semantics (JWS 2020) [pdf]  
  • Shrestha Ghosh, Simon Razniewski, Gerhard Weikum. CounQER: A System for Discovering and Linking Count Information in Knowledge Bases. (ESWC 2020) [pdf] [demo] [poster]

 

Demos and Posters

Exploring count information in Wikidata and DBpedia.
Answering entity count queries from search snippets.
  • CoQEx: user demonstration for the entity count queries.
  • CounQER: user demonstration on our predicate alignment queries on Wikidata and DBpedia. 

Datasets and Code

  • Class cardinality comparison: [data|code]
  • Answering count queries with explanations: [data|code]