Interlinking unstructured text information with model-based project data: an approach to product model based information mining

摘要

Even with an increasing integration of model-based systems with project information spaces, a large percentage of the business and engineering knowledge in AEC/FM is still captured implicitly in isolated and often poorly structured text documents. In this paper we propose a Bayesian Network based mining framework utilising product model data as a primary source of engineering knowledge to support domain-specific information retrieval as well as re-organisation and re-contextualisation of document information. Central to the framework is a four layer Bayesian Network enabling to combine methods for analysing text documents and filtering product models in an evidential reasoning process. Capturing, combining and visualising available analyses results as well as the current mining context, the network allows for explicitly representing the knowledge on document repositories in semantic content networks according to a user's task and information needs. Furthermore, we expect the mining network, explicitly interlinking document and product model information, to support the understanding of the interrelations among the two domains as well as the exploration of new retrieval, mining and integration approaches.