Rob Gaizauskas: Research Interests

Areas of Research Interest/Activity

• Information extraction and text mining from natural language texts, especially newspaper and biomedical texts
• Temporal and spatial information in natural language -- representation, annotation and extraction
• Automatic image description
• Argument mining and summarisation from multiparty conversations
• Automatic recognition and analysis of scene structure in narrative texts
• Dialogue to support spoken language interaction with information sources
• Reasoning and Large Language Models
• Comparable corpora -- acquisition and use (e.g. for machine translation)
• Single and multi-document summarization
• Evaluation and NLP
• Software Architectures for NLP

Current and Recently Funded Research Projects

• Text Correction for Historical Documents

Previous Funded Research Projects

• A Multimodal Speech and Graphical Interface for Hands-free Data Capture and Querying in MRO
• Investigating Spoken Dialogue to Support Manufacturing Processes
• SENSEI -- Making Sense of Human-Human Conversation
• ViSen -- Visual Sense. Tagging visual data with semantic descriptions
• Accurat -- Analysis and Evaluation of Comparable Corpora for Under Resourced Areas of Machine Translation
• BioWSD -- The Biomedical Word Sense Disambiguation Project
• CLARITY -- Interactive Cross Language Retrieval
• CLEF -- Clinical E-Science Framework
• Cub Reporter -- Question Answering and Summarisation for Gathering Background
• ECRAN -- Extraction of Content: Research at Near Market
• ELSE -- Evaluation in Language and Speech Engineering
• EMILLE -- Enabling Minority Language Engineering
• EMPathIE -- Enzyme and Metabolic Pathways Information Extraction
• GATE -- A General Archictecture for Text Engineering
• GATE-2 -- A General Archictecture for Text Engineering -- Version 2
• GoTag -- Real-time Text Mining for the Biomedical Literature: A Collaboration between DiscoveryNet and myGrid
• KDisc -- Knowledge Discovery from the Biomedical Literature
• LaSIE -- Large Scale Information Extraction
• MeTeR -- Measuring Text Reuse
• MyGrid -- Directly Supporting the E-Scientist
• PASTA -- Protein Active Site Template Acquistion -- Large Scale Information Extraction System
• POETIC -- The POrtable Extendable Traffic Information Collator
• SToBS -- Structured Transcription of Broadcast Speech
• TaaS -- Terminology as a Service
• TRESTLE -- Text Retrieval, Extraction and Summarization for Large Enterprises
• TRIPOD -- Automating Caption Creation
• VIKEF -- Virtual Information and Knowledge Environment Framework