Knowledge-Assisted Retrieval of Online Product Information in A/E/C (Architecture/engineering/construction)

128 p.Thesis (Ph.D.)--University of Illinois at Urbana-Champaign, 2005.Product cost is a significant expenditure in a construction project where project participants have had to search for and dealt with a large amount of product information. With more product information, industry practitioners can make more informed project decisions. However, current approaches that support A/E/C information acquisition do not provide a comprehensive market scan. While the Internet provides an ever-growing resource for product information, none of the existing approaches is able to make use of this virtual market. Therefore, this research has developed a knowledge-assisted approach to tackle issues related to online searching to help retrieve A/E/C product information from the Internet. The research goal particularly is to increase the number of distinct manufacturers identified for a given product. The research investigated issues related to online product information searching and reviewed related literatures to identify tools for handling these issues. Then, a research model was built accordingly utilizing domain knowledge, information retrieval (IR) techniques, and strategies that incorporate domain knowledge into knowledge-supported IR approaches. Specifically, domain knowledge represented in the form of a thesaurus was used with query expansion strategies under the framework of an adaptation of the extended Boolean model. A prototype was implemented with five data sets for research validation. It was concluded from the testing results that domain knowledge can be helpful and the developed approach can effectively increase the number of distinct product manufacturers identified.U of I OnlyRestricted to the U of I community idenfinitely during batch ingest of legacy ETD