Iterative semi-automatic modularization of documents and topic map creation for product development knowledge
Harald Weber, Marko Lenhart, Herbert Birkhofer · TUbilio (Technical University of Darmstadt) · 2007
The aim of the current pinngate-project [1] at the product development and machine elements (pmd) department at the Darmstadt University of Technology is a flexible knowledge system for the learning, teaching and application of product development knowledge. Close interaction between the cognitive and the information technical part is essential for its success. The first specifies the requirements for the contents of documents, while the second deals with the modularization of the documents. Both are taken into account by the creation of a semantic net for advanced information retrieval. The problem is finding an appropriate methodology for this task. While manual methods are very work intensive and time intensive, fully automatic methods do not normally provide the required quality. A compromise is a semi-automatic procedure, which supports the user in preparing a document for modularization, textual analyses and integrating it in the system by means of several tools. Further analyses also help the user to identify weak points in the structure of the original documents. The information gained can then be used to improve the documents and their modularization.