Joint learning of named entity recognition and relation extraction

Named entities are important to extract relations. Accurate relation classification helps recognize named entities. The paper presents a joint approach of named entity recognition and relation identification. The identified relation is utilized to improve named entity recognition. The method has been applied to identify the names of persons and organizations and five relations between them. The result shows that the joint approach has improved the recall and F-measure of named entities without scarifying the precision. Meanwhile, the recall and F-measure are also improved in the relation extraction.