Technical Perspective: Entity Matching with Quality and Error Guarantees

Abstract

The challenge of entity matching is that of identifying when different data items (often referred to as records or mentions) refer to the same real-life entity. Popular instantiations of this problem include deduplication, where the items are database records that include duplicate representations of the same entity (e.g., duplicate profiles in a social network) [2], record linkage, where the items come from different data sources that mention overlapping sets of entities (e.g., the profiles of two social networks) [5], and schema matching, where the items are attributes of different database schemas that intersect on their domain of interest (e.g., the database schemas of different social networks) [6].