Guiding Theme A1: Entity Linking/Cross-document Coreference Resolution

The NLP task entity linking links mentions of entities to concepts in a knowledge base which is derived from Wikipedia (such as Yago or DBpedia). This task is difficult, because mentions can be highly ambiguous (just check how many different Wikipedia pages the mention king has), and this page does not even include all the people with the last name King), and because the same concept can be referred to by many different mentions. The related task of cross-document coreference resolution does not link mentions to concepts but identifies whether mentions in different documents refer to the same entity. This task is also difficult because the mention "King" in one document can refer to the same entity as the mention "King" in another document, or not. AIPHES source documents are from various genres, which makes the tasks of entity linking and cross-document coreference resolution more difficult as mentions of entities may vary considerably across genres.


PhD. Projects may address entity linking or cross-document coreference resolution, but preferably both together, since the tasks are related and may support each other. This can be solved via graph- or network-based methods or advanced machine learning techniques for jointly modeling both tasks. The participation at shared tasks like the entity linking task at the Text Analysis Conference is encouraged.


Poster (in German)

Example thesis topics


  • Entity linking for German
  • Genre- and domain-independent concept disambiguation and -clustering
  • Using graphs to integrate concept disambiguation, concept disambiguation and coreference resolution


A A A | Drucken Print | Impressum Impressum | Sitemap Sitemap | Suche Search | Kontakt Contact | Webseitenanalyse: Mehr Informationen
zum Seitenanfangzum Seitenanfang