Optimizing Product Design through a Particle Swarm Induced Logistic Regression Model

This paper defines how a meta heuristic search engine called p.s.o can be used to maximize the objective function of a logistic regression model, describing the relationship between the response variable (product designs' score) and a set of explanatory variables (product design factors). at the first phase the processed data, classified and categorized, by kansie engineering is used as input to the logistic regression model. the pso optimization algorithm, maximizes the likelihood function, thus the parameters of the model, being the coefficients of the independent factors are estimated. after post hoc tests, the validated model, defines the relation between consumer semantic and physical product selection factors and the response variable which is a dichotomous dependent ordered variable representing product design scores.