A knowledge-aware graph-based framework for document-level event extraction

doi:10.1016/j.knosys.2026.116942

A knowledge-aware graph-based framework for document-level event extraction

2026-09-04

0

OA

AI

DOI：10.1016/j.knosys.2026.116942

Original

Original request for help

Save

Abstract

En 中文

• Proposes a knowledge-aware framework for document-level event extraction. • Models documents as relational graphs to capture structured knowledge. • Introduces pseudo-trigger combination for robust event structure identification. • Enhances long-range dependency modeling across sentences and entities. • Enables transformation of unstructured text into structured event knowledge.

Keywords:

Knowledge extraction

Document-level event extraction

Knowledge representation

Graph neural networks

Information extraction

A knowledge-aware graph-based framework for document-level event extraction

Abstract

Journal

Organization

Cited Papers