In the vast majority of Wikipedia article, the noun phrase at the beginning of the article is the name of the entity described by the article it-self, even though there is no self pointing link to referencing it-self. This make the NER corpus scripts miss many potentially informative links that might hurt the performance of the trained OpenNLP models.
In the vast majority of Wikipedia article, the noun phrase at the beginning of the article is the name of the entity described by the article it-self, even though there is no self pointing link to referencing it-self. This make the NER corpus scripts miss many potentially informative links that might hurt the performance of the trained OpenNLP models.