NetGraph SystemSearching through the Prague Dependency Treebank

The goal of one of the projects being solved at our Center is to build a corpus of Czech with a rich annotation scheme–the Prague Dependency Treebank. Text files containing these data in the form of labeled trees are being created automatically with manual corrections. With the increasing amount of annotated material searching a corpus becomes a complex task. The present paper describes the software system that was developed to contribute to the solution of this problem. Some requirements were established during the analysis of the design: the software has to be able to work in the Internet environment, to display tree structures in a graphical form, and the client part has to be hardware independent.