Yoshio Kajitani

Learn More
The forecasting capabilities of Feed-Forward Neural Network (FFNN) models are compared to those of other competing time series models by carrying out forecasting experiments. As demonstrated by the detailed forecasting results for the Canadian lynx data set, FFNN models perform very well, especially when the series contains nonlinear and non-Gaussian(More)
This paper introduces the methodological challenge of identifying and quantifying the interdependencies among several critical infrastructures. First, interdependency structures during a natural disaster are modelled-based on past events, considering supply (electricity, water and gas), communication (internet and telephone) and transportation(More)
This study, based on a questionnaire survey and workshops, and with a focus on the impact of an earthquake on the Nagata Elementary School Community in Kobe City, Japan, develops a collaborative model to assess the allocation of residents to shelters. The current official allocation plan is compared with three alternative allocations developed within the(More)
Spatial temporal GIS named DiMSIS (Disaster Management Spatial Information System) have been developed mainly focused on applying local government applications after the great HANSHIN earthquake of 1995 in Japan. This spatial temporal GIS which has been enhanced to have multi language support would be common core system to support Risk Adaptive Regional(More)
  • 1