Relationship extraction model for the cybersecurity domain

Project description

This project aims to build a relationship extraction model for the cybersecurity domain. Relationship extraction involves extracting meaningful associations between concepts and entities from the textual content. Any relationship extraction module aims to identify textual segments that indicate semantic relationships between entities. These relationships fall into several types or classes, such as friend_of, child_of, married_to, located_in, etc. Relationship extraction is vital in building knowledge graphs or semantic networks where nodes (concepts) are connected to other nodes via relationships. In the context of cybersecurity, the relation extraction system would be able to extract relationship mentions from the text about cybersecurity concepts such as attacks, vulnerabilities, threats, etc. The trained cyber security relationship extraction model should generate semantic triples resulting in a structured representation of textual content, which a security analyst can use to inspect and respond to a cyber security attack.

Co-supervisors

Dr Saeed Rehman

Assumed knowledge

Understanding of the cybersecurity domain

Note: You need to register interest in projects from different supervisors (not a number of projects with the one supervisor).
You must also contact each supervisor directly to discuss both the project details and your suitability to undertake the project.