Taxonomy Expansion for Named Entity Recognition

TL;DR

This paper introduces the Partial Label Model (PLM), a novel method for expanding NER taxonomies using partially annotated data, significantly reducing annotation effort while improving recognition performance especially with limited data.

Contribution

The paper presents PLM, a new approach for taxonomy expansion in NER that outperforms existing methods using only partially annotated datasets, reducing annotation costs.

Findings

PLM outperforms most approaches by 0.5-2.5 F1 points.

PLM is especially effective with limited data for new entity types.

Taxonomy expansion becomes more cost-effective with PLM.

Abstract

Training a Named Entity Recognition (NER) model often involves fixing a taxonomy of entity types. However, requirements evolve and we might need the NER model to recognize additional entity types. A simple approach is to re-annotate entire dataset with both existing and additional entity types and then train the model on the re-annotated dataset. However, this is an extremely laborious task. To remedy this, we propose a novel approach called Partial Label Model (PLM) that uses only partially annotated datasets. We experiment with 6 diverse datasets and show that PLM consistently performs better than most other approaches (0.5 - 2.5 F1), including in novel settings for taxonomy expansion not considered in prior work. The gap between PLM and all other approaches is especially large in settings where there is limited data available for the additional entity types (as much as 11 F1), thus suggesting a more cost effective approaches to taxonomy expansion.