Résumé
Today's assistive technologies aim to improve the quality of information acquisition for visually impaired persons. Screen readers and screen magnifiers are the main visual assis-tive technologies that are currently proposed. However, due to economical reasons, these technologies do not consider specific needs of visually impaired people. In this paper, we propose an approach to make Web pages more accessible for users who have specific needs. User wishes, like text color, font size, link color or even more complex wishes including wishes about brightness or contrast are encoded as user preferences. We also encode designer graphical choices as designer preferences. From these preferences, a new Web page design is computed using a resolution algorithm. We compare two distinct approaches to make this computation: resolution algorithms from preference theory domain and an evolutionary algorithm. The comparison evaluates running time on automatically generated data and on the sign in Facebook page.