Neural Information Extraction on Technical Short Text: From Theory to Practical Systems

A collection of software and demos for my thesis entitled "Neural Information on Technical Short Text: From Theory to Practical Systems".

Chapter 3: E2EET: From Mention-level to End-to-End Entity Typing

The E2EET model described in the chapter is available at https://github.com/Michael-Stewart-Webdev/e2e-entity-typing. The repository also includes the three datasets (Modified BBN, Modified Ontonotes, and Wiki_50k) described in the chapter.

Chapter 4: Domain-independent Knowledge Graph Construction

The Seq2KG system is available here: https://github.com/Michael-Stewart-Webdev/Seq2KG. The repository also includes the three datasets (CateringServices, AutomotiveEngineering, and BBN) described in the chapter.

Chapter 5: Lexical Normalisation

The Targeted model is available at https://github.com/Michael-Stewart-Webdev/lexnorm.

The General model is available at https://github.com/Michael-Stewart-Webdev/pytorch-lexnorm.

The US Accidents dataset is available at https://github.com/Michael-Stewart-Webdev/us-accidents-dataset.

Chapter 6: Aquila: An Interactive Web-based Toolset for Knowledge Discovery from Short Text

Aquila is available at https://github.com/Michael-Stewart-Webdev/aquila.

A live demo of Aquila is currently running at https://nlp-tlp.org/aquila.

Chapter 7: Redcoat: A Collaborative Annotation Tool for Hierarchical Entity Typing

Redcoat is available at https://github.com/Michael-Stewart-Webdev/redcoat.