View in EDS

Document level Relationship Extraction based on context feature enhancement.

Saved in:
Bibliographic Details
Title: Document level Relationship Extraction based on context feature enhancement.
Authors: Zhang, Nan1 (AUTHOR), Cui, Ziming1 (AUTHOR), Cai, Qiang1 (AUTHOR)
Source: Pattern Recognition Letters. Nov2025, Vol. 197, p24-30. 7p.
Subject Terms: *CONTEXTUAL analysis, *FEATURE extraction, *LANGUAGE models, *DATA quality, *DATA mining
Abstract: Document level Relationship Extraction (DocRE) tasks aim to extract relationships between multiple entities from long texts. However, obtaining feature representations for entity pairs that span multiple sentences is a challenge. Additionally, the feature information for triplets depends on both intra-document and inter-sentence information. To address this issue, this paper proposes a model named Plus-DocRE for DocRE(PDRE). Firstly, we introduce entity segmentation based on spans to increase the potential number of entities and improve negative sample recognition. Secondly, we utilize the BERT pre-trained model to obtain paragraph and local context information, enriching the features of entity pairs. Finally, through linear layers and self-attention mechanisms, we fuse the features of local and paragraph context for multi-label relationship classification, enabling entity relationship extraction. Meanwhile, we introduce a new data mechanism, C-DocRE, to simulate a more realistic scenario with annotation errors. Experimental results show that the PDRE model outperforms other baseline models in performance, achieving an F1 score of 53.6. • Introducing paragraph level contextual attention to enhance the representation of entity pairs. • Expanding the range of negative samples using span based entity negative sample expansion method. • Fusion of local and paragraph contextual features of entity pairs through feature fusion methods. [ABSTRACT FROM AUTHOR]
Database: Academic Search Index
Description
Abstract:Document level Relationship Extraction (DocRE) tasks aim to extract relationships between multiple entities from long texts. However, obtaining feature representations for entity pairs that span multiple sentences is a challenge. Additionally, the feature information for triplets depends on both intra-document and inter-sentence information. To address this issue, this paper proposes a model named Plus-DocRE for DocRE(PDRE). Firstly, we introduce entity segmentation based on spans to increase the potential number of entities and improve negative sample recognition. Secondly, we utilize the BERT pre-trained model to obtain paragraph and local context information, enriching the features of entity pairs. Finally, through linear layers and self-attention mechanisms, we fuse the features of local and paragraph context for multi-label relationship classification, enabling entity relationship extraction. Meanwhile, we introduce a new data mechanism, C-DocRE, to simulate a more realistic scenario with annotation errors. Experimental results show that the PDRE model outperforms other baseline models in performance, achieving an F1 score of 53.6. • Introducing paragraph level contextual attention to enhance the representation of entity pairs. • Expanding the range of negative samples using span based entity negative sample expansion method. • Fusion of local and paragraph contextual features of entity pairs through feature fusion methods. [ABSTRACT FROM AUTHOR]
ISSN:01678655
DOI:10.1016/j.patrec.2025.07.006