Ambiguity recognition and nocuous ambiguity

Almost all text is ambiguous in some way. This research theme investigates how humans interpret ambiguous text, and how ambiguity can be either recognised or resolved by computers. This research supports a range of applications in which the ability to interpret ambiguous text is important, especially automatic assessment, interpreting requirements documents, and text mining from social media.

Key publications

• Understanding RT’s Audiences: Exposure Not Endorsement for Twitter Followers of Russian State-Sponsored Media.
• Mapping networks of influence: tracking Twitter conversations through time and spac
• Using NLP to support scalable assessment of short free text responses
• A generalised hybrid architecture for NLP
• Analysing anaphoric ambiguity in natural language requirement