Rough set and PSO-based ANFIS approaches to modeling customer satisfaction for affective product design

摘要: Rough set and PSO-based ANFIS approaches are proposed to model customer satisfaction in affective product design.Rough is used simplify structure PSO introduced enhance the accuracy of modeling.Generated fuzzy rules address fuzziness exists survey data.Generated models show nonlinear relationships between responses design attributes.The outperform FLSR, FR, GP-FR terms training errors validation errors. Facing fierce competition marketplaces, companies try determine optimal settings attribute new products from which best can be obtained. To settings, relating customers attributes have first developed. Adaptive neuro-fuzzy inference systems (ANFIS) was attempted previous research shown an effective approach data nonlinearity modeling for design. However, incapable that involve a number inputs may cause failure process lead 'out memory' error. overcome limitation, this paper, rough (RS) particle swarm optimization (PSO) based-ANFIS further improve accuracy. In approaches, RS theory adopted extract significant as employed parameter explicit with better generated. A case study mobile phones illustrate approaches. The results based on compared those ANFIS, least-squares regression (FLSR), (FR), genetic programming-based (GP-FR). Results tests perform than others