We are now accepting applications for a 3-year PhD studentship on Knowledge Modelling, Discovery and Reasoning with focus on Natural Language Processing in the School of Computing Science and Digital Media, Robert Gordon University
The aim of the project is to explore the hypothesis that partial or complete knowledge graphs provide necessary focus and context for interactive and dynamic conversational design.
Creating these graphs can be a manual burden. Therefore, the project will study state-of-the-art NLP and NLU for domain-specific Named-Entity Recognition (NER) and relation extraction to enable interactive discovery and population of knowledge graphs.
Ontologies and knowledge graphs capture knowledge in the form of entities and relationships between them, and are being used in applications across domains such as geosciences, healthcare, retail, and manufacturing. A partially created knowledge graph also provides an opportunity to contextualise conversational interactions between the machine and the user. Other advantages of incorporating such knowledge within a real-world reasoning tasks include: providing a formal definition of the data used by reasoning mechanisms; easing the sharing of that data by semantically describing it; and reducing the effort involved with integrating data from third parties by providing a data vocabulary/dictionary for describing the external data.
In this industrial collaborative project, you will be exposed to the application of ontologies within reasoning platforms intended to address real-world challenges. In particular, you will study the consequences of using a network of ontologies and associated knowledge graphs to inform conversational interaction with users. You will also explore the extraction of named-entities from user responses to deliver contextually relevant recommendations via reasoning methodologies such as case-based reasoning.
The project will also investigate related problems such as similarity metrics for extraction and annotation; testing the feasibility of the approach on real-world industry data; and synthesizing suitable language generation or automatically providing concise explanations for the reasoning.
- Masters degree with Distinction (or a First Class honours BSc in exceptional cases) from a UK university or equivalent overseas/professional qualification in computer science, artificial intelligence, mathematics, or other related subject.
- Strong analytical and object-oriented and/or functional programming and software engineering skills.
- Knowledge or experience in at least one field of symbolic AI (such as analogical reasoning, case-based reasoning, ontological modelling and reasoning, etc.).
- Good understanding or knowledge of at least two of the following:
- Information extraction
- NLP or NLG
- Deep Learning for NLP
- Conversational AI
- Excellent programming skills.
- Good command of English in writing and speaking.
- Ability to work collaboratively in a team, willingness to explore areas related to the target domain and excellent inter-personal skills
The studentship covers full time PhD tuition fees on an international or home basis, and a tax free stipend of GBP £14,777 per year for 3 years.
Applications should be emailed to Kate Lines at CSDM-ResearchAdmin@rgu.ac.uk by Wednesday 4th March 2020.
The applications should consist of a covering letter or personal statement of interest, and CV. Further information such as passport details or transcripts may be requested during the short-listing stage. All shortlisted applicants will be interviewed before an offer is made.