In natural language processing, Entity Linking, also referred to as named-entity disambiguation (NED), named-entity recognition and disambiguation (NERD) or named-entity normalization (NEN)[1] is the task of assigning a unique identity to entities (such as famous individuals, locations, or companies) mentioned in text. For example, given the sentence "Paris is the capital of France", the main idea is to first identify "Paris" and "France" as named entities, and then to determine that "Paris" refers to the city of Paris and not to Paris Hilton or any other entity that could be referred to as "Paris" and "France" to the french country. The Entity Linking task is composed of 3 subtasks. First, Named Entity Recognition, which consist in the extraction of named entities from a text. Second, for each named entity, the objective is to generate candidates from a Knowledge Base (e.g. Wikipedia, Wikidata, DBPedia, ...). We call this step candidate generation. The main challenge being that we want to get the corresponding entity inside the candidates set. Lastly, the objective is to choose from the candidate set the correct entity. We call this step disambiguation.
On way to see the Entity Linking task is as the combination of the NER task followed by the NEL task, the NEL task being the combination of candidate generation and disambiguation.
