Abstract

The FrameNet database contains descriptions of more than 7,000 lexical units based on more than 130,000 annotated sentences. The database and its related software are central to the process of entering lexical information, annotating sentences, displaying the results, and distributing the FrameNet data. This article discusses both how the design of the database follows the principles of frame semantics and also how the database provides appropriate access for project staff and users elsewhere.

Author notes

1International Computer Science Institute, Berkeley, California, USA 2International Computer Science Institute and University of California, Berkeley, California, USA 3International Computer Science Institute, Berkeley, California and Massachusetts Institute of Technology, USA