Radboud University at TREC 2020

The IR research group from the Radboud University (RUIR) has an interest in creating graph-based solutions for domain specific challenges. We aim to increase effectiveness by incorporating domain knowledge into graph development. This work was developed as part of a Master’s thesis project about graph representations for news articles of TREC’s background linking task [1]. The introduced work explores the use of named entities, novelty scores and diversification of background documents.