GraphIE: A Graph-Based Framework for Information Extraction
arXiv:1810.13083
Abstract
Most modern Information Extraction (IE) systems are implemented as sequential taggers and only model local dependencies. Non-local and non-sequential context is, however, a valuable source of information to improve predictions. In this paper, we introduce GraphIE, a framework that operates over a graph representing a broad set of dependencies between textual units (i.e. words or sentences). The algorithm propagates information between connected nodes through graph convolutions, generating a richer representation that can be exploited to improve word-level predictions. Evaluation on three different tasks --- namely textual, social media and visual information extraction --- shows that GraphIE consistently outperforms the state-of-the-art sequence tagging model by a significant margin.
arXiv commentNAACL 2019
References in corpus (6)
- Adam: A Method for Stochastic Optimization
- Semi-Supervised Classification with Graph Convolutional Networks
- Character-Aware Neural Language Models
- Harnessing Deep Neural Networks with Logic Rules
- End-to-End Relation Extraction using LSTMs on Sequences and Tree Structures
- Graph Convolution over Pruned Dependency Trees Improves Relation Extraction
Cited by in corpus (11)
- Graph Convolutional Networks for Graphs Containing Missing Features
- Graph Neural Networks for Natural Language Processing: A Survey
- FreeDOM: A Transferable Neural Architecture for Structured Information Extraction on Web Documents
- PICK: Processing Key Information Extraction from Documents using Improved Graph Learning-Convolutional Networks
- Hierarchical Contextualized Representation for Named Entity Recognition
- Abstractive Information Extraction from Scanned Invoices (AIESI) using End-to-end Sequential Approach
- Robust Layout-aware IE for Visually Rich Documents with Pre-trained Language Models
- Interpretable Multi-dataset Evaluation for Named Entity Recognition
- KARL-Trans-NER: Knowledge Aware Representation Learning for Named Entity Recognition using Transformers
- Web Question Answering with Neurosymbolic Program Synthesis
- Parameter-Efficient Neural Question Answering Models via Graph-Enriched Document Representations