An Ontology-Based Intelligent Information System for Urbanism and Civil Engineering Data Stefan Trausan-Matu 1,2 and Anca Neacsu 1 1 “Politehnica” University of Bucharest, Department of Computer Science and Engineering, Splaiul Independetei nr. 313, Bucharest, Romania 2 Research Institute for Artificial Intelligence of the Romanian Academy Calea 13 Septembrie nr.13, Bucharest, Romania Corresponding author: Stefan Trausan-Matu, email: trausan@cs.pub.ro fax: +40.318.153290 phone: +40.724.985518 Abstract. The paper presents a prototype of an intelligent information system for urban and civil engineering data centered on an ontology. The system will provide intelligent personalized access to concepts from the ontology and to a collection of relevant associated documents indexed according to the ontology’s concepts. The declarative knowledge of the ontology and an associated set of production rules may be used for automatic inferences that will enable reasoning for getting intelligent answers to users’ queries, under an expert system dialog. The paper uses examples from a first version of the ontology for urban and civil engineering concepts that is in the center of the system. The ontology is developed following an integrated cognitive and socio-cultural approach. It contains a taxonomy of objects structured according to Engestrom’s Theory of Activity and Sowa’s ontology. This structured development facilitates further knowledge acquisition. Keywords: information systems, ontology, Semantic Web, expert systems, production rules, Jess 1 Introduction The number of documents containing data about urbanism and civil engineering is becoming higher every day. Moreover, some of them are changing as new recommendations, regulations, and laws appear or substitute old ones. The majority of this data is now available on the web and they may be accessed both by general search engines like Google (http://www.google.com) or by specific search tools available on particular web sites. However, a major problem is that all these search engines are keyword-based and they are not able to make inferences and to cope with relationships among documents. 75 Ontologies for urban development: conceptual models for practitioners