Skip to content

pwichmann/phd_thesis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 

Repository files navigation

Towards automatically generating supply chain maps from natural language text

About

This repository contains supplementary material for my PhD thesis "Towards automatically generating supply chain maps from natural language text". This repository shall also provide pointers to the academic papers that have been written as part of this project.

Content

The material contains the following:

  • Distinction between supply chain and related concepts
  • Supply chain resilience
  • A critical reflection on the value of supply chain maps
  • Inter- and intra-annotator agreement
  • Use cases of structural supply chain visibility

Publications

The following publications resulted from the research conducted as part of the PhD.

PhD thesis "Towards automatically generating supply chain maps from natural language text" (2021)

Conference paper "Towards automatically generating supply chain maps from natural language text" (2018)

Alexandra Brintrup, Simon Baker, Philip Woodall, and Duncan McFarlane.Towards automatically generating supply chain maps from natural language text. IFAC-PapersOnLine, 51(11):1726 – 1731, 2018. ISSN 2405-8963. 16th IFAC Symposium on Information Control Problems in Manufacturing INCOM 2018.

Link to the paper

https://www.sciencedirect.com/science/article/pii/S2405896318313284

You can also contact the author here (create an 'Issue' in this repo) to get access to the manuscript.

Journal paper "Extracting supply chain maps from news articles using deep neural networks" (2020)

Pascal Wichmann, Alexandra Brintrup, Simon Baker, Philip Woodall & Duncan McFarlane (2020) Extracting supply chain maps from news articles using deep neural networks, International Journal of Production Research, 58:17, 5320-5336, DOI: 10.1080/00207543.2020.1720925

Link to the paper

https://www.tandfonline.com/doi/abs/10.1080/00207543.2020.1720925

You can also contact the author here (create an 'Issue' in this repo) to get access to the manuscript.

Supplementary material

Please see here for material supplementary to the IJPR paper and code examples: https://github.com/pwichmann/supply_chain_mining

Industrial application and commercialisation

Versed AI is a new Cambridge University spin-out being formed by a post-doc and post-graduates from several different university departments. The team has developed Natural Language Processing (NLP) and Machine Learning (ML) technology for business intelligence purposes.

The technology can text-mine millions of news articles, business reports and social media for relationships between organisations, companies, products, and people. This information can be used to create vast knowledge networks that artificial intelligence is applied to in order to discover patterns and infer missing or unknown knowledge. These networks have many useful applications including predicting future relations and discovering information from the network structure. The first application the team intends to focus on is to automatically extract supply chain maps.

Among other achievements, the Versed AI team has won the Entrepreneurial Postdocs of Cambridge business plan competition (including a £20k investment sponsored by Cambridge Enterprise).

Releases

No releases published

Packages

No packages published