Inverse Analysis of Aesthetic Evaluation of Planted Concrete Structures by Neural Networks
Yasuo Chikata, Noboru Yasuda, Manabu MATSUSHIMA, Tameo Kobori · Computer-Aided Civil and Infrastructure Engineering · 1998
Scenery evaluation depends on the evaluator's experiences or perceptions. Thus inverse analysis by neural networks of scenery evaluation of planted concrete structures (concrete retaining walls) was examined in this study. In addition, in order to identify the evaluation schemes of evaluators, sensitivity analysis was performed on the obtained neural network structure. The efficacy of neural network inverse analysis and genetic algorithm analysis using fuzzy-set theory in reproduction of the same scenery evaluation questionnaire's results also was compared.