Building a Cyber Security Named Entity Recognition and Classification System

Project description

This project aims to build a named entity recognition model for the cybersecurity domain. Named entity recognition and classification is an essential component of information extraction systems. It plays a vital role in natural language processing tasks such as machine translation, summarization, and knowledge extraction. It identifies text segments from a textual document corresponding to real-world entities. It is a two-step process where in the first step, it identifies the named entity type, and in the second step, it detects the boundary of the recognized entity. Existing NER models work great for generic entities such as people, places, and organizations but are inappropriate for cyber security domains. Therefore, a NER model is required to effectively identify and classify cyber security entities like security concepts, attacks, countermeasures, vulnerabilities, threats, tools, etc. This project will create a state-of-the-art classifier for cyber security NER that can identify named entities in different security forums, blogs, articles, news, etc.

Co-supervisors

Dr Saeed Rehman

Assumed knowledge

Understanding of cyber security 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.