TL;DR本文基于图形依存解析思想,利用双仿射模型对输入进行全局视图,探索所有语句中的跨度,以准确预测命名实体,以 Nested NER 为例, 对 8 个语料库进行了评估并取得了所有语料库的 SOTA 性能,最高的准确度增益高达 2.2 个百分点。
Abstract
named entity recognition (NER) is a fundamental task in Natural Language
Processing, concerned with identifying spans of text expressing references to
entities. NER research is often focused on flat entities only (flat NER),
ignoring the fact that entity references can be nested, as in