Applying non-monotonic reasoning to Natural Language Generation
Natural Language Generation concerns itself with the translation of formal knowledge into natural language (e.g. English). Several highly successful techniques for this process have been identified, and this project concerns itself with investigating ways of combining these techniques to improve outputs even further.
The project will investigate the application of a blackboard-like system driven by a set of non-monotonic rules to the generation of natural language. More specifically, each different technique might generate slightly different text, with different levels of clarity, brevity, etc. Each of these levels will provide a different level of justification as to why it should be used, and the system will aggregate the different texts, via rules, to provide some optimal text (based on user preferences). A user can then query the system to understand the reasons why specific rules were used, and cause different texts to be generated by providing justifications as to why specific rules are invalid.
This project lies at the unexplored boundary of two mature areas of research, namely natural language generation and argumentation.
The successful applicant should have, or expect to have, an Honours Degree at 2.1 or above (or equivalent) in Computing Science or a mathematical discipline (e.g. maths, physics, applied maths)
There is no funding attached to this project, it is for self-funded students only.
Formal applications can be completed online: http://www.abdn.ac.uk/postgraduate/apply. You should apply for PhD in Computing Science, to ensure that your application is passed to the correct College for processing. Please ensure that you quote the project title and supervisor on the application form.
Informal inquiries can be made to Dr N Oren ([email protected]) with a copy of your curriculum vitae and cover letter. All general enquiries should be directed to the Graduate School Admissions Unit ([email protected]).