Abstract
Most of the evaluation problems are comprehensive and with ever-increasingly more uncertainties. By quantifying the involved uncertainties, Basic Uncertain Information can both well handle and merge those uncertainties in the input information. This study proposed a two-level comprehensive evaluation model by using some merging techniques which can consider both the original preference information and the bi-polar preference over the information with high certainty degrees. A numerical application in educational evaluation is also proposed to verify the effectiveness and flexibility of the proposed model. (C) 2021 The Authors. Published by Atlantis Press B.V.