Modeling Designers' Color Decision Processes Through Emotive Choice Mapping

摘要

Color selection support systems require a quantitative model of the color design decision-making process in order to support color selection strategies that further the specified goals of the designer without obstructing the unspecified goals. The system described in this paper models the color selection decision process based on the current state of the design, the desired state of the design, which is based on specified and unspecified designer goals. The specified goals are quantified as subjective responses to completed designs. In the main study discussed, seven novice designers independently designed 20 web pages and, in the process, every color selection was recorded. Adjective pairs selected from monologues provided semantic differential for evaluation of the resulting designs. A neural network-based inference system models designer selection based on the eventual results and the current state of the design to form designer strategies for color design support. This research is relevant in a variety of interactive applications but is of special interest for systems that work in close conjunction with human creativity.