Start of funding 01.01.2015

Treebank@NewsScape

Dr. Peter Uhrig
Friedrich-Alexander-University of Erlangen-Nuremberg
Institut für Anglistik und Amerikanistik

Prof. Dr. Francis F. Steen
University of California, Los Angeles (UCLA)
Department of Communication Studies



The proposed collaboration will lay the groundwork for a joint grant proposal for developing next-generation search tools for the UCLA NewsScape Library of International Television News, a multimodal and multilingual corpus of 250,000 hours of television news, annotated by more than three billion words and categories with a suite of cutting-edge tools in linguistics and computer vision. The search engine will be based on the treebank.info project, which was developed to make it possible to search in grammatical structures that are represented as unrestricted graphs. It has completed development of a user-friendly web interface and a powerful scalable backend made available as open source software. A prototype of the extended search software will be developed during the project and presented to the public at selected conferences.