This paper presents a first version of LinkPeople, an entity-centric system for coreference resolution of person entities. The approach combines (ⅰ) a multi-pass architecture which takes advantage of entity features at document-level with (ⅱ) a set of linguistically-motivated constraints and rules which allows the system to restrict the candidates of a given mention. The paper includes evaluations and error analysis of LinkPeople in 3 different languages, achieving promising results (more than 81% F1 in different metrics). Both the system and the corpora are freely distributed.
展开▼
