Deep Joint Entity Disambiguation with Local Neural Attention

2017/04/17 by Ganea, Octavian-Eugen, Hofmann, Thomas · 2 citations
#Computation and Language (cs.CL) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.1704.04920

Abstract

We propose a novel deep learning model for joint document-level entity disambiguation, which leverages learned neural representations. Key components are entity embeddings, a neural attention mechanism over local context windows, and a differentiable joint inference stage for disambiguation. Our approach thereby combines benefits of deep learning with more traditional approaches such as graphical models and probabilistic mention-entity maps. Extensive experiments show that we are able to obtain competitive or state-of-the-art accuracy at moderate computational costs.

Cited by

• An Entity Linking Agent for Question Answering
• Effective Multi-Task Learning for Biomedical Named Entity Recognition

Related

• Joint Learning of the Embedding of Words and Entities for Named Entity Disambiguation
• Leveraging Deep Neural Networks and Knowledge Graphs for Entity Disambiguation
• End-to-End Neural Entity Linking
• Contextualized End-to-End Neural Entity Linking
• Distributed Entity Disambiguation with Per-Mention Learning