it.cnr.isti.hpc.dexter.disambiguation

Interface Disambiguator

All Known Implementing Classes:

TopScoreEntityDisambiguator

public interface Disambiguator

A disambiguator is one of the main component of an entity tagger. A spot could refer to more than one entity, for example consider the sentence:

On July 20, 1969, the Apollo 11 astronauts - Neil Armstrong, Michael Collins, and Edwin “Buzz” Aldrin Jr. - realized President Kennedy’s dream.

It is quite easy to map the spot to the entity John F. Kennedy, since in Wikipedia there are 98 anchors exactly matching such fragment of text and linking to the U.S. president page. On the other hand, it could be not so easy for a software to decide if Michael Collins is an astronaut, an Irish leader or the president of the Irish provisional government in 1922.

Given a list of matched spots, where each spot has one or more candidate entities, a disambiguator produces a list of entities, so that each spot can have at most one entity.

Author:

Diego Ceccarelli Created on Aug 2, 2013

Method Summary

Methods

Modifier and Type	Method and Description
EntityMatchList	disambiguate(SpotMatchList sml) Given a list of matched spots, where each spot has one or more candidate entities, returns a list of entities, so that each spot can have at most one entity.

Method Detail

disambiguate

EntityMatchList disambiguate(SpotMatchList sml)

Given a list of matched spots, where each spot has one or more candidate entities, returns a list of entities, so that each spot can have at most one entity.

Parameters:

sml - - the list of spots detected in a document, each spot match contains a list of candidate entities.