Xu Xiuqin, Xie Jialiang, Wang Honghui, Lin Mingwei
College of Science, Jimei University, Xiamen, Fujian 361021 China.
College of Mathematics and Finance, Putian University, Putian, Fujian 351100 China.
Appl Intell (Dordr). 2022;52(12):13659-13674. doi: 10.1007/s10489-022-03289-7. Epub 2022 Mar 5.
During the COVID-19, colleges organized online education on a massive scale. To make better use of online education in the post-epidemic era, this paper conducts an online education satisfaction survey with four types of colleges and 129,325 students propose a fuzzy TOPSIS (technique for order preference by similarity to ideal solution) method based on the cloud model to rank the satisfaction of different colleges. Firstly, based on the characteristics of online education during the COVID-19, we build an evaluation indicator system from four dimensions: technology, instructor, learner and environment including, 10 indicators and 94 sub-indicators. Secondly, the cloud model is used to quantitatively describe the natural language and uncertainty in a large amount of assessment information. The cloud model generator is used for sub-indicators and achieves an effective and flexible conversion between linguistic information and quantitative values. The cloud model of indicators are presented by integrating the corresponding sub-indicators. The weights of indicators are determined by the entropy method based on the cloud model and possibility degree matrix, which eliminates the judgment of decision-makers and has great power for handling practical problems with unknown weight information. Finally, a fuzzy TOPSIS method based on the cloud model is proposed to rank the satisfaction of online education of different colleges. The proposed method is compared with other existing methods to shown its merits. The experimental result is consistent with the proportion of students who accept online education in the post-epidemic era. According to the second questionnaire, as the qualitative evaluation of the cloud model of indicators increases, the qualitative evaluation of satisfaction of different types of colleges will also increase. It indicates that the method proposed in this paper is practical.
在新冠疫情期间，高校大规模组织了在线教育。为了在后疫情时代更好地利用在线教育，本文对四类高校和129325名学生进行了在线教育满意度调查，提出了一种基于云模型的模糊TOPSIS（逼近理想解排序法）方法来对不同高校的满意度进行排名。首先，基于新冠疫情期间在线教育的特点，从技术、教师、学习者和环境四个维度构建了一个评价指标体系，包括10个指标和94个子指标。其次，利用云模型对大量评估信息中的自然语言和不确定性进行定量描述。云模型生成器用于子指标，实现了语言信息和定量值之间的有效灵活转换。通过整合相应子指标得到指标的云模型。基于云模型和可能性度矩阵，利用熵权法确定指标权重，消除了决策者的主观判断，对于处理权重信息未知的实际问题具有很强的能力。最后，提出了一种基于云模型的模糊TOPSIS方法对不同高校的在线教育满意度进行排名。将所提方法与其他现有方法进行比较以展示其优点。实验结果与后疫情时代接受在线教育的学生比例一致。根据第二次问卷调查，随着指标云模型定性评价的增加，不同类型高校满意度的定性评价也会增加。这表明本文提出的方法具有实用性。
