A Joint Named Entity Recognition and Entity Linking System

We present a joint system for named entity recognition (NER) and entity linking (EL), allowing for named entities mentions extracted from textual data to be matched to uniquely identifiable entities. Our approach relies on combined NER modules which transfer the disambiguation step to the EL component, where referential knowledge about entities can be used to select a correct entity reading. Hybridation is a main feature of our system, as we have performed experiments combining two types of NER, based respectively on symbolic and statistical techniques. Furthermore, the statistical EL module relies on entity knowledge acquired over a large news corpus using a simple rule-base disambiguation tool. An implementation of our system is described, along with experiments and evaluation results on French news wires. Linking accuracy reaches up to 87%, and the NER fmeasure up to 83%.