CHAPTER 4

Relation Extraction

4.1    INTRODUCTION

Relation extraction (RE) is the task of extracting links between entities, which builds on the task of NERC discussed in the previous chapter. The focus is often to extract binary relations between named entities, but can also be to extract more complex relations, such as events. Typical relation types include birthdate(PER, DATE) and founder-of(PER, ORG), with examples for relations being birthdate(John Smith, 1985-01-01) or founder-of(Bill Gates, Microsoft).

RE can be an annotation task, i.e., to annotate a text with relations, but is typically considered a slot filling task, also called knowledge base population, i.e., to populate a knowledge base with relations for a given set of relation ...

Get Natural Language Processing for the Semantic Web now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.