作者: H. Li , L. Y. Shen , P. E. D. Love
DOI: 10.1061/(ASCE)0733-9364(1999)125:3(185)
关键词:
摘要: Artificial neural networks (ANNs) have been applied to support construction mark-up estimation. The major drawback of this application, however, is that an ANN system unable explain why and how a particular recommendation made. This significantly affects the user-acceptance its results. research presented in paper investigates use KT-1 method for automatically extracting rules from trained network. implemented tested on collected bidding data, results investigation indicate usefulness method. Discussions difficulties generating automated explanations are also presented.