GREG was an INTAS Georgia project between the Intelligent Systems Group of the Computer Science Department at the University of Stuttgart (coordinator), the Information Technology Research Institute at the University of Brighton, the Computational Lexicography Group, Department of Applied Mathematics and Computer Science at the Tbilisi State University, and the Institute of Linguistics, Georgian Academy of Sciences.

The goal of the GREG project was to develop a multilingual valency lexicon that is suitable for use in various NLP-applications. It contains contain syntactic and semantic valency patterns for about 1000 verbal equivalents from English, Georgian, German and Russian. Syntactic valency patterns are specified in terms of subcategorization frames; semantic valency patterns draw on the thematic (or functional) roles as introduced by Halliday, Chafe, Longacre and others.

Both the micro and the macro structure of the lexicon follow the general principles for an efficient representation of multilingual valency lexica that have also been developed in the project. The representation formalism used to encode the GREG lexicon is DATR.

• Citation forms covered in GREG (to read Georgian and Russian data, you need these fonts)
  
• Lists of subcategorization frames used in GREG
  
• Subcat (and partly case frame) lexica (English, German, Russian)
• German verbal valency hierarchies ( move, mental, verbal, transfer, communication)