Scaling Trust: Veracity-Driven Defect Detection in Entity Search

Veracity is a critical dimension of data quality that directly impacts a wide range of tasks. In entity search scenarios, Knowledge Graphs (KGs) such as DBpedia and Wikidata serve as core resources for accessing factual content. The veracity of these KGs is therefore essential for ensuring the reliability and trustworthiness of retrieved entities - factors that directly influence user confidence in the search system. However, ensuring the truthfulness of entities remains a major challenge due to the complexities associated with the scale, development, and maintenance of KGs. This paper critically analyzes the impact of veracity in entity search, using DBpedia as the underlying KG. To this end, we introduce eRank, a veracity-driven re-ranking strategy that enhances entities' trustworthiness without sacrificing the ranking's overall relevance. Furthermore, we propose the Active Learning-based verAcity-Driven Defect IdentificatioN (ALADDIN) system, a lightweight and scalable framework for veracity-driven defect detection. ALADDIN identifies incorrect KG facts and exhibits high effectiveness in downstream entity-centric tasks, such as entity summarization, entity card generation, and defect recommendation.