This book is about syntactic databases (a.k.a. treebanks), collections of text material in which the syntactic relations have been made visible. It starts off with a general intro-duction to the subject and then continues with three in-depth investigations of more specialized aspects.
In the introduction, syntactic databases are first placed in the larger context of linguistic databases, text collections with a broader linguistic annotation than just a syntactic one. Then some examples of syntactic databases are given, illustrating the range of annotation actually encountered. The introduction is completed with an investigation of database management systems for syntactic databases.
The first in-depth investigation concerns the treatment of ambiguous structures in syntactic analysis trees, focussing on a very efficient representation for such structures and the means to create this representation. Next, classroom use of syntactic databases is examined. A computer program for this purpose, CLUES, is discussed, along with a suggested series of syntax exercises. The final subject is the importance of including function and attribute information in the annotation of texts. The central line of investigation here is a probabilistic parsing experiment in which the use of function and attribute information is the main variable.