This paper proposes a cloud multi-criteria group decision-making model for teacher evaluation in higher education which is involving subjectivity, imprecision and fuzziness. First, selecting the appropriate evaluation index depending on the evaluation objectives, indicating a clear structural relationship between the evaluation index and objectives and establishing a proper evaluation system are all critical and fundamental tasks. Then, collect expert evaluation data, process data, establish training set to build the decision trees, extract evaluation rules, simplify evaluation process, while reducing the cost of evaluation in real applications. Third, establish the interval cloud evaluation matrix through the decision cloud, transforming the evaluation value through the cloud model, determining the order of importance of the decision program, and make the decision. Finally, an addressing linguistic decision-making problem for college teacher evaluation is provided to illustrate the effectiveness of the proposed model.