Empirical Study of Named Entity Recognition Performance Using Distribution-aware Word Embedding

TL;DR

This paper presents an empirical study on how distribution-aware word embeddings, which incorporate word specificity, can improve Named Entity Recognition performance across different document types.

Contribution

It introduces a novel distribution-aware word embedding method and demonstrates its effectiveness in enhancing NER accuracy by leveraging word specificity information.

Findings

Distribution-aware embeddings improve NER performance

Incorporating specificity enhances detection of unfamiliar entities

Method outperforms baseline NER models

Abstract

With the fast development of Deep Learning techniques, Named Entity Recognition (NER) is becoming more and more important in the information extraction task. The greatest difficulty that the NER task faces is to keep the detectability even when types of NE and documents are unfamiliar. Realizing that the specificity information may contain potential meanings of a word and generate semantic-related features for word embedding, we develop a distribution-aware word embedding and implement three different methods to make use of the distribution information in a NER framework. And the result shows that the performance of NER will be improved if the word specificity is incorporated into existing NER methods.