RWTH Aachen University of Technology
Department of Computer Science III
52074 Aachen, Germany
The CHASID project investigates semantical modeling of informational texts (such as scientific articles or textbooks). By constructing the model at the time of writing, additional support can be provided for the author. The aim is a kind of ``semantical spelling checker'' - spotting common suboptimal structures, but without taking guesses at the meaning of text. Instead, the model is provided completely by the author. Easing this burden is a central task in the project. As a solution, instantiable substructures, consisting of parts of the model and document template text, have been introduced. By recording the instances and the importances of their part, an easily understandable way to define consistency checks has been found.
This has been implemented in a prototype using PROGRES and UPGRADE, connecting with ToolBook and Emacs psgml.
For more information, see a three-page English summary
Created by: Svenlast modification: Monday 06 of December, 2004 [13:18:52 UTC] by