Text structure recognition using a region algebra

We consider the problem of incrementally developing a parser for text structure. This means building the parser specification a piece at a time while simultaneously developing our understanding of the text. We argue that existing solutions have usability and efficiency problems for this application and propose an alternative approach based on the type of region algebra that is often used as a query language for text databases. This is an appropriate interface for incremental development, but has no efficient batch parsing model such as those that exist for grammars. In this thesis, we propose an efficient batch parsing model and characterize the region algebras to which it applies.