Deep Neural Approaches to Relation Triplets Extraction: A Comprehensive Survey

TL;DR

This survey reviews recent deep neural network methods for relation triplet extraction, covering various architectures, datasets, and emerging directions like zero-shot learning, to guide future research in the field.

Contribution

It provides a comprehensive overview of deep learning approaches for relation extraction, highlighting recent advancements, datasets, and future research directions.

Findings

Deep neural models achieve state-of-the-art performance.

Coverage of diverse neural architectures and datasets.

Identification of emerging research trends like zero-shot extraction.

Abstract

Recently, with the advances made in continuous representation of words (word embeddings) and deep neural architectures, many research works are published in the area of relation extraction and it is very difficult to keep track of so many papers. To help future research, we present a comprehensive review of the recently published research works in relation extraction. We mostly focus on relation extraction using deep neural networks which have achieved state-of-the-art performance on publicly available datasets. In this survey, we cover sentence-level relation extraction to document-level relation extraction, pipeline-based approaches to joint extraction approaches, annotated datasets to distantly supervised datasets along with few very recent research directions such as zero-shot or few-shot relation extraction, noise mitigation in distantly supervised datasets. Regarding neural architectures, we cover convolutional models, recurrent network models, attention network models, and graph convolutional models in this survey.