Coreference Resolution in Portuguese: Detecting Person, Location and Organization

Abstract

Coreference resolution is a task of great relevance for Natural Language Processing area, given that the performance of many other tasks depends on the correct output of this type of system, especially the extraction of relationships between named entities. The present work aims at resolving coreference in Portuguese, focusing on the following categories of named entities: Person, Location and Organization. The proposed method uses supervised learning. To this end, the selection and implementation of features that assist in the correct classification are fundamental, since the classification model is built from this data. KeywordsCoreference Resolution; Natural Language Processing; Named Entities, Machine Learning.