Abstract
The purpose of this paper is to provide the method for the deduction of impression of streets in historic district of Kyoto City by using Self-Organizing Maps (SOM). The street data consisted of spatial elements given from the analysis. SOM includes a two-dimensional visible system which allowed us to analyze the multivariate data. By using SOM, we were able to create a map reflecting the characteristics of the streets with the impression by sensibility evaluation. In this case the characteristics and the sensibility evaluation data of the streets were easily understood by visual analysis. By executing SOM on the new additional data of a street consisted of spatial elements, we deduced the impression of historical streets and the proposal for improvement of spatial elements based on the similarity of them by the location on this map
