Kansei Knowledge-Based Human-Centric Digital Interface Design Using BP Neural Network

Digital interface has increasingly replaced the traditional human–computer hardware interface and become the main carrier of human–computer interaction in information intelligent system. How to design and develop an effective digital interface is a new problem faced by enterprises and designers. Aiming at the practical problems of cognitive difficulties such as overload and mismatch in the field of digital interface design of complex information systems, this paper proposed a method for human-centric digital interface design based on Kansei knowledge. It was done to study the Kansei knowledge of digital interface to determine the Kansei images that affects the interface, identify the key elements of interface design including interface layout style, main color style, font style, and core component expression, and then construct a nonlinear mapping and mathematical prediction model between the Kansei images and elements of interface design based on BP neural network. Finally, the feasibility of this method was verified, which can effectively match the user’s specific perceptual cognitive needs of complex digital interface.