Automated Classification of Construction Project Documents

Abstract

The number of documents generated in a construction project and stored in interorganizational information systems is significant. Since a large percentage of these project documents are generated in text format, methods for organizing and improving access to the information contained in these types of documents become essential to construction information management. Information classification schemes can be used for this purpose. They provide a common framework to enact document organization and information exchange among project members. Current systems for document management rely on manual classification methods controlled by human experts. Due to the widespread use of information technologies for construction, the increasing availability of electronic documents, and the development of systems based on project object models, manual classification becomes unfeasible. This paper presents a unique way to improve information organization and access in interorganizational systems based on automated classification of construction project documents according to their related project components. Machine learning methods were used for this purpose. A prototype of a document classification system was developed to provide easy deployment and scalability to the classification process.