A Feature-Based Model for Nested Named-Entity Recognition at VLSP-2018 NER Evaluation Campaign

TL;DR

This paper presents a feature-based sequence labeling approach for nested named-entity recognition, demonstrating improved accuracy by jointly tagging multiple entity levels in the VLSP-2018 dataset.

Contribution

The study introduces a joint-tag model that effectively handles nested entities by combining multiple features and entity levels, advancing nested NER techniques.

Findings

Joint-tag model improves nested NER accuracy

Feature combination enhances recognition performance

Effective handling of nested entities in sequence labeling

Abstract

In this report, we describe our participant named-entity recognition system at VLSP 2018 evaluation campaign. We formalized the task as a sequence labeling problem using BIO encoding scheme. We applied a feature-based model which combines word, word-shape features, Brown-cluster-based features, and word-embedding-based features. We compare several methods to deal with nested entities in the dataset. We showed that combining tags of entities at all levels for training a sequence labeling model (joint-tag model) improved the accuracy of nested named-entity recognition.