2006 Volume 12 Issue 23 Pages 449-452
The purpose of this study is to develop the optimum design method for roof shapes satisfying both beauty and dynamic condition by using rough sets. At the previous study stage, neural network was used for learning KANSEI. There are parts of the black box in neural network. The relation between roof shapes and KANSEI is not clear. For this problem solving, rough sets was used instead of neural network. By analyzing the feature of roof shapes and KANSEI by using rough sets, even if it is not shape of the same condition, KANSEI evaluation is possible.
Ethical standards of the publisher (‘Ethical Standards Concerning the Peer Review for the Presentation of Architectural Transactions and Works’)