Abstract
This paper presents a neural network based approach to modeling consumers' affective responses for product form design. A theoretical framework for a single user's perception is developed. On the basis of this theoretical framework, a mathematical model which enables single users' responses to different products to be predicted was developed. The results obtained show that the mathematical models developed achieved highly accurate predictions.
For the purpose of obtaining a global model various individual mathematical models were created, which were based on the opinions of users representing different groups of opinion. The results suggest that, under some conditions, the combined use of various models of individual users can perform as well as a single model generated on the basis of mean market responses.
