Dependency-Guided LSTM-CRF for Named Entity Recognition

IJCNLP 2019  ·  Zhanming Jie, Wei Lu ·

Dependency tree structures capture long-distance and syntactic relationships between words in a sentence. The syntactic relations (e.g., nominal subject, object) can potentially infer the existence of certain named entities. In addition, the performance of a named entity recognizer could benefit from the long-distance dependencies between the words in dependency trees. In this work, we propose a simple yet effective dependency-guided LSTM-CRF model to encode the complete dependency trees and capture the above properties for the task of named entity recognition (NER). The data statistics show strong correlations between the entity types and dependency relations. We conduct extensive experiments on several standard datasets and demonstrate the effectiveness of the proposed model in improving NER and achieving state-of-the-art performance. Our analysis reveals that the significant improvements mainly result from the dependency relations and long-distance interactions provided by dependency trees.

PDF Abstract IJCNLP 2019 PDF IJCNLP 2019 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Named Entity Recognition (NER) CoNLL 2003 (English) DGLSTM-CRF + ELMo (L=2) 3.0pt1-4.51.5 F1 92.4 # 40
Chinese Named Entity Recognition OntoNotes 5.0 DGLSTM-CRF F1 79.92 # 1
Named Entity Recognition (NER) Ontonotes v5 (English) DGLSTM-CRF + ELMo F1 89.88 # 15
Named Entity Recognition (NER) Ontonotes v5 (English) DGLSTM-CRF (L=2) F1 88.52 # 18

Methods


No methods listed for this paper. Add relevant methods here