Establishing formalized representation of specifications for construction cost estimation by using ontology learning

Construction cost estimation for tendering is important for both tenders and bidders in construction projects and needs to be strictly complied with corresponding specifications so that quantities and prices from different bidders are comparable. In the view of flexibility and extensibility, ontology is regarded as a promising technology for formalized representation of the specifications for construction cost estimation (cost specifications for short hereafter) in computer programs. Because, with the support of reasoning machines for ontology, it is possible to automatize the processes of construction cost estimation. However, the manual establishment of ontology for construction cost estimation (cost ontology for short hereafter) is labor-intensive and time-consuming, not to mention that there are numbers of specifications for different types of construction projects in different region. Thus, the cost ontologies are hampered from practical use. In order to solve this problem, a semi-automatic approach based on the framework that authors established previously is proposed to establishing the formalized representation of cost specifications, i.e. the cost ontology, using ontology learning technology. Firstly, the data sources, i.e. the cost specifications are analyzed and the corresponding relations between information in cost specifications and the elements in the framework of cost ontology are summarized. Then based on the corresponding relations, the approach is designed, in which concepts, relations and rules are extracted by natural language processing and domain lexical analysis to fill the framework of cost ontology. The approach lays a foundation for the practical use of ontology in construction cost estimation.