Research For Supporting Joint Venture Using Organization Information On Web
Masanori Ikebe, Shigenori Tanaka, Hitoshi Furuta, Kenji D. Nakamura, Kenta Kobayashi · Proceedings of the Joint CIB W78, W102, ICCCBE, ICCC, and DMUCE International Conference on Computing and Decision Making in Civil and Building Engineering, Montreal, Canada, 14-16 June · 2006
Today, the case has increased that some company organizes JV and takes part in the construction works because it’s difficult for a single company to correspond to a large-scale construction works. However, there’s a problem that a lot of JV have only been organized among the companies connected at the past. This is the reason that each company cannot share the reliable information such as its technology, capital and so on with each other. Then, this paper suggests how to analyze the organization information from the Web and extract the reliable relations among the companies. We suggest the Web structure analysis that combined link structure analysis with the natural language processing in order to collecting the reliable information. We adopt the improved HITS algorithm as the link structure analysis. And, in the natural language processing, we extract any topic information using the morphological analysis and analyze the information related to the Web pages without link relations. Then, in order to confirm the suggested method available, we compare the result of the method with the fact that is united JV using F measure that is used to judge the precision of the extracted information. We confirm that the method is available.