Awesome Relation Extraction

A curated list of awesome resources dedicated to Relation Extraction, inspired by awesome-nlp and awesome-deep-vision.

Contributing: Please feel free to make pull requests.

Contents

• Research Trends and Surveys
• Papers
  • Supervised Approaches
  • Distant Supervision Approaches
  • GNN-based Models
  • Language Models
    • Encoder Representation from Transformer
    • Decoder Representation from Transformer
  • Knowledge Graph Based Approaches
  • Few-Shot Learning Approaches
• Datasets
• Videos and Lectures
• Systems
• Frameworks

Research Trends and Surveys

• NLP progress: Relationship Extraction
• Named Entity Recognition and Relation Extraction:State-of-the-Art (Nasar et al., 2021)
• A Survey of Deep Learning Methods for Relation Extraction (Kumar, 2017)
• A Survey on Relation Extraction (Bach and Badaskar, 2017)
• Relation Extraction: A Survey (Pawar et al., 2017)
• A Review on Entity Relation Extraction (Zhang et al., 2017)
• Review of Relation Extraction Methods: What is New Out There? (Konstantinova et al., 2014)
• 100 Best Github: Relation Extraction

Papers

Supervised Approaches

CNN-based Models

• Convolution Neural Network for Relation Extraction [paper] [code] [review]
  • ChunYang Liu, WenBo Sun, WenHan Chao and WanXiang Che
  • ADMA 2013
• Relation Classification via Convolutional Deep Neural Network [paper] [code] [review]
  • Daojian Zeng, Kang Liu, Siwei Lai, Guangyou Zhou and Jun Zhao
  • COLING 2014
• Relation Extraction: Perspective from Convolutional Neural Networks [paper] [code] [review]
  • Thien Huu Nguyen and Ralph Grishman
  • NAACL 2015
• Classifying Relations by Ranking with Convolutional Neural Networks [paper] [code]
  • Cicero Nogueira dos Santos, Bing Xiang and Bowen Zhou
  • ACL 2015
• Attention-Based Convolutional Neural Network for Semantic Relation Extraction [paper] [code]
  • Yatian Shen and Xuanjing Huang
  • COLING 2016
• Relation Classification via Multi-Level Attention CNNs [paper] [code]
  • Linlin Wang, Zhu Cao, Gerard de Melo and Zhiyuan Liu
  • ACL 2016
• MIT at SemEval-2017 Task 10: Relation Extraction with Convolutional Neural Networks [paper]
  • Ji Young Lee, Franck Dernoncourt and Peter Szolovits
  • SemEval 2017

RNN-based Models

• Relation Classification via Recurrent Neural Network [paper]
  • Dongxu Zhang and Dong Wang
  • arXiv 2015
• Bidirectional Long Short-Term Memory Networks for Relation Classification [paper]
  • Shu Zhang, Dequan Zheng, Xinchen Hu and Ming Yang
  • PACLIC 2015
• End-to-End Relation Extraction using LSTMs on Sequences and Tree Structure [paper]
  • Makoto Miwa and Mohit Bansal
  • ACL 2016
• Attention-Based Bidirectional Long Short-Term Memory Networks for Relation Classification [paper] [code]
  • Peng Zhou, Wei Shi, Jun Tian, Zhenyu Qi, Bingchen Li, Hongwei Hao and Bo Xu
  • ACL 2016
• Semantic Relation Classification via Hierarchical Recurrent Neural Network with Attention [paper]
  • Minguang Xiao and Cong Liu
  • COLING 2016
• Semantic Relation Classification via Bidirectional LSTM Networks with Entity-aware Attention using Latent Entity Typing [paper] [code]
  • Joohong Lee, Sangwoo Seo and Yong Suk Choi
  • arXiv 2019

Dependency-based Models

• Semantic Compositionality through Recursive Matrix-Vector Spaces [paper] [code]
  • Richard Socher, Brody Huval, Christopher D. Manning and Andrew Y. Ng
  • EMNLP-CoNLL 2012
• Factor-based Compositional Embedding Models [paper]
  • Mo Yu, Matthw R. Gormley and Mark Dredze
  • NIPS Workshop on Learning Semantics 2014
• A Dependency-Based Neural Network for Relation Classification [paper]
  • Yang Liu, Furu Wei, Sujian Li, Heng Ji, Ming Zhou and Houfeng Wang
  • ACL 2015
• Classifying Relations via Long Short Term Memory Networks along Shortest Dependency Path [paper] [code]
  • Xu Yan, Lili Mou, Ge Li, Yunchuan Chen, Hao Peng and Zhi Jin
  • EMNLP 2015
• Semantic Relation Classification via Convolutional Neural Networks with Simple Negative Sampling [paper]
  • Kun Xu, Yansong Feng, Songfang Huang and Dongyan Zhao
  • EMNLP 2015
• Improved Relation Classification by Deep Recurrent Neural Networks with Data Augmentation [paper]
  • Yan Xu, Ran Jia, Lili Mou, Ge Li, Yunchuan Chen, Yangyang Lu and Zhi Jin
  • COLING 2016
• Bidirectional Recurrent Convolutional Neural Network for Relation Classification [paper]
  • Rui Cai, Xiaodong Zhang and Houfeng Wang
  • ACL 2016
• Neural Relation Extraction via Inner-Sentence Noise Reduction and Transfer Learning [paper]
  • Tianyi Liu, Xinsong Zhang, Wanhao Zhou, Weijia Jia
  • EMNLP 2018

GNN-based Models

• Matching the Blanks: Distributional Similarity for Relation Learning [paper]
  • Livio Baldini Soares, Nicholas FitzGerald, Jeffrey Ling, Tom Kwiatkowski
  • ACL 2019
• Relation of the Relations: A New Paradigm of the Relation Extraction Problem [paper]
  • Zhijing Jin, Yongyi Yang, Xipeng Qiu, Zheng Zhang
  • EMNLP 2020
• GDPNet: Refining Latent Multi-View Graph for Relation Extraction [paper] [code]
  • Fuzhao Xue, Aixin Sun, Hao Zhang, Eng Siong Chng
  • AAAI 21
• RECON: Relation Extraction using Knowledge Graph Context in a Graph Neural Network [parer] [code]
  • Anson Bastos, Abhishek Nadgeri, Kuldeep Singh, Isaiah Onando Mulang', Saeedeh Shekarpour, Johannes Hoffart, Manohar Kaul
  • WWW'21

Distant Supervision Approaches

• Distant supervision for relation extraction without labeled data [paper] [review]
  • Mike Mintz, Steven Bills, Rion Snow and Dan Jurafsky
  • ACL 2009
• Knowledge-Based Weak Supervision for Information Extraction of Overlapping Relations [paper] [code]
  • Raphael Hoffmann, Congle Zhang, Xiao Ling, Luke Zettlemoyer and Daniel S. Weld
  • ACL 2011
• Multi-instance Multi-label Learning for Relation Extraction [paper] [code]
  • Mihai Surdeanu, Julie Tibshirani, Ramesh Nallapati and Christopher D. Manning
  • EMNLP-CoNLL 2012
• Distant Supervision for Relation Extraction via Piecewise Convolutional Neural Networks [paper] [review] [code]
  • Daojian Zeng, Kang Liu, Yubo Chen and Jun Zhao
  • EMNLP 2015
• Relation Extraction with Multi-instance Multi-label Convolutional Neural Networks [paper] [review] [code]
  • Xiaotian Jiang, Quan Wang, Peng Li, Bin Wang
  • COLING 2016
• Incorporating Relation Paths in Neural Relation Extraction [paper] [review]
  • Wenyuan Zeng, Yankai Lin, Zhiyuan Liu and Maosong Sun
  • EMNLP 2017
• Neural Relation Extraction with Selective Attention over Instances [paper] [code]
  • Yankai Lin, Shiqi Shen, Zhiyuan Liu, Huanbo Luan and Maosong Sun
  • ACL 2017
• Learning local and global contexts using a convolutional recurrent network model for relation classification in biomedical text [paper] [code] [code]
  • Desh Raj, Sunil Kumar Sahu and Ashish Anan
  • CoNLL 2017
• Hierarchical Relation Extraction with Coarse-to-Fine Grained Attention[paper][code]
  • Xu Han, Pengfei Yu∗, Zhiyuan Liu, Maosong Sun, Peng Li
  • EMNLP 2018
• RESIDE: Improving Distantly-Supervised Neural Relation Extraction using Side Information [paper] [code]
  • Shikhar Vashishth, Rishabh Joshi, Sai Suman Prayaga, Chiranjib Bhattacharyya and Partha Talukdar
  • EMNLP 2018
• Distant Supervision Relation Extraction with Intra-Bag and Inter-Bag Attentions [paper] [code]
  • Zhi-Xiu Ye, Zhen-Hua Ling
  • NAACL 2019

Language Models

Encoder Representation from Transformer

• Enriching Pre-trained Language Model with Entity Information for Relation Classification [paper]
  • Shanchan Wu, Yifan He
  • arXiv 2019
• LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention [paper] [code]
  • Ikuya Yamada, Akari Asai, Hiroyuki Shindo, Hideaki Takeda, Yuji Matsumoto
  • EMNLP 2020
• SpanBERT: Improving pre-training by representing and predicting spans [paper] [code]
  • Mandar Joshi, Danqi Chen, Yinhan Liu, Daniel S. Weld, Luke Zettlemoyer and Omer Levy
  • TACL 2020 (Transactions of the Association for Computational Linguistics)
• Efficient long-distance relation extraction with DG-SpanBERT [paper]
  • Jun Chen, Robert Hoehndorf, Mohamed Elhoseiny, Xiangliang Zhang

Decoder Representation from Transformer

• Improving Relation Extraction by Pretrained Language Representations [paper] [review] [code]
  • Christoph Alt, Marc Hübner, Leonhard Hennig
  • AKBC 19

Knowledge Graph Based Approaches

• KGPool: Dynamic Knowledge Graph Context Selection for Relation Extraction [paper] [code]
  • Abhishek Nadgeri, Anson Bastos, Kuldeep Singh, Isaiah Onando Mulang, Johannes Hoffart, Saeedeh Shekarpour, and Vijay Saraswat
  • ACL 2021 (findings)

Few-Shot Learning Approaches

• FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation [paper] [website] [code]
  • Xu Han, Hao Zhu, Pengfei Yu, Ziyun Wang, Yuan Yao, Zhiyuan Liu, Maosong Sun
  • EMNLP 2018

Miscellaneous

• Jointly Extracting Relations with Class Ties via Effective Deep Ranking [paper]
  • Hai Ye, Wenhan Chao, Zhunchen Luo and Zhoujun Li
  • ACL 2017
• End-to-End Neural Relation Extraction with Global Optimization [paper]
  • Meishan Zhang, Yue Zhang and Guohong Fu
  • EMNLP 2017
• Adversarial Training for Relation Extraction [paper]
  • Yi Wu, David Bamman and Stuart Russell
  • EMNLP 2017
• A neural joint model for entity and relation extraction from biomedical text[paper]
  • Fei Li, Meishan Zhang, Guohong Fu and Donghong Ji
  • BMC bioinformatics 2017
• Joint Extraction of Entities and Relations Using Reinforcement Learning and Deep Learning [paper]
  • Yuntian Feng, Hongjun Zhang, Wenning Hao, and Gang Chen
  • Journal of Computational Intelligence and Neuroscience 2017
• TDEER: An Efficient Translating Decoding Schema for Joint Extraction of Entities and Relations [paper] [code]
  • Xianming Li, Xiaotian Luo, Chenghao Dong, Daichuan Yang, Beidi Luan and Zhen He
  • EMNLP 2021

Back to Top

Datasets

• SemEval-2010 Task 8 [paper] [download]
  • Multi-Way Classification of Semantic Relations Between Pairs of Nominals
• New York Times (NYT) Corpus [paper] [download]
  • This dataset was generated by aligning Freebase relations with the NYT corpus, with sentences from the years 2005-2006 used as the training corpus and sentences from 2007 used as the testing corpus.
• FewRel: Few-Shot Relation Classification Dataset [paper] [Website]
  • This dataset is a supervised few-shot relation classification dataset. The corpus is Wikipedia and the knowledge base used to annotate the corpus is Wikidata.
• TACRED: The TAC Relation Extraction Dataset [paper] [Website] [download]
  • Is a large-scale relation extraction dataset with built over newswire and web text from the corpus used in the yearly TAC Knowledge Base Population (TAC KBP) challenges.
• ACE05: [Website] [download-info]
  • This dataset represent texts extracted from a variety of sources: broadcast conversation, broadcast news, newsgroups, weblogs. The 6 relation types between 7 types on entities: acility (FAC), Geo-PoliticalEntity (GPE), Location (LOC), Organization (ORG), Person (PER), Vehicle (VEH), Weapon (WEA).
• SemEval-2018 Task 7 [paper] [Website] [download]
  • The corpus is collected from abstracts and introductions of scientific papers, and there are six types of semantic relations in total. There are three subtasks of it: Subtask 1.1 and Subtask 1.2 are relation classification on clean and noisy data, respectively; Subtask 2 is the standard relation extraction.

For state of the art results check out nlpprogress.com on relation extraction

Back to Top

Videos and Lectures

• Stanford University: CS124, Dan Jurafsky
  • (Video) Week 5: Relation Extraction and Question
• Washington University: CSE517, Luke Zettlemoyer
  • (Slide) Relation Extraction 1
  • (Slide) Relation Extraction 2
• New York University: CSCI-GA.2590, Ralph Grishman
  • (Slide) Relation Extraction: Rule-based Approaches
• Michigan University: Coursera, Dragomir R. Radev
  • (Video) Lecture 48: Relation Extraction
• Virginia University: CS6501-NLP, Kai-Wei Chang
  • (Slide) Lecture 24: Relation Extraction

Back to Top

Systems

• DeepDive
• Stanford Relation Extractor

Back to Top

Frameworks

• OpenNRE [github] [paper]
  • Is an open-source and extensible toolkit that provides a unified framework to implement neural models for relation extraction (RE) between named entities. It is designed for various scenarios for RE, including sentence-level RE, bag-level RE, document-level RE, and few-shot RE. It provides various functional RE modules based on both TensorFlow and PyTorch to maintain sufficient modularity and extensibility, making it becomes easy to incorporate new models into the framework.
• AREkit [github] [research-applicable-paper]
  • Is an open-source and extensible toolkit focused on data preparation for document-level relation extraction organization. It complements the OpenNRE functionality, as in terms of the latter, document-level RE setting is not widely explored (2.4 [paper]). The core functionality includes (1) API for document presentation with EL (Entity Linking, i.e. Object Synonymy) support for sentence level relations preparation (dubbed as contexts) (2) API for contexts extraction (3) relations transferring from sentence-level onto document-level, etc. It provides neural networks (like OpenNRE) and BERT modules, both applicable for sentiment attitude extraction task.
• DeRE [github] [paper]
  • Is an open-source framework for declaritive relation extraction, and therefore allows to declare your own task (using XML schemas) and apply manually implemented models towards it (using a provided API). The task declaration builds on top of the spans and relations between spans. In terms of the latter, authors propose frames, where every frame yelds of: (1) trigger (span) and (2) n-slots, where every slot may refer to frame or span. The framework poses no theoretical restrictions to the window from which frames are extracted. Thus, this concept may cover sentence-level, document-level and multi-document RE tasks.

Back to Top

License

To the extent possible under law, Joohong Lee has waived all copyright and related or neighboring rights to this work.