A Cascade Dual-Decoder Model for Joint Entity and Relation Extraction

TL;DR

This paper introduces a cascade dual-decoder model that jointly extracts overlapping entities and relations from text, improving accuracy in knowledge graph construction tasks, especially with complex overlapping triples.

Contribution

The paper presents a novel cascade dual-decoder approach that jointly models entity and relation extraction, effectively handling overlapping triples and reducing error propagation.

Findings

Achieved higher F1 scores on real-world and public datasets.

Demonstrated effectiveness and generalizability of the method.

Outperformed traditional pipeline methods in extracting complex triples.

Abstract

In knowledge graph construction, a challenging issue is how to extract complex (e.g., overlapping) entities and relationships from a small amount of unstructured historical data. The traditional pipeline methods are to divide the extraction into two separate subtasks, which misses the potential interaction between the two subtasks and may lead to error propagation. In this work, we propose an effective cascade dual-decoder method to extract overlapping relational triples, which includes a text-specific relation decoder and a relation-corresponded entity decoder. Our approach is straightforward and it includes a text-specific relation decoder and a relation-corresponded entity decoder. The text-specific relation decoder detects relations from a sentence at the text level. That is, it does this according to the semantic information of the whole sentence. For each extracted relation, which is with trainable embedding, the relation-corresponded entity decoder detects the corresponding head and tail entities using a span-based tagging scheme. In this way, the overlapping triple problem can be tackled naturally. We conducted experiments on a real-world open-pit mine dataset and two public datasets to verify the method's generalizability. The experimental results demonstrate the effectiveness and competitiveness of our proposed method and achieve better F1 scores under strict evaluation metrics. Our implementation is available at https://github.com/prastunlp/DualDec.