Teaching Quality Assessment of English Courses in Colleges and Universities Based on ISSA-DRNN

Abstract

Abstract The optimized sparrow search algorithm (ISSA) is built on top of the deep neural network (DNN) in this paper, along with a new deep recurrent neural network (DRNN) algorithm that is proposed by improving the DNN. Ultimately, an ISSA-DRNN-based model for assessing the quality of English course teaching has been established. To examine the application impact of the ISSA-DRNN-based English course quality teaching evaluation model, SA University has been chosen as the study location. Two groups have been created: an experimental group and a control group. The experimental group’s academic level scores grew by 5.1 points in terms of English achievement, whereas the control group’s average score increased by 0.36 points, indicating a very significant difference (P<0.01). The four characteristics of satisfaction with teaching effectiveness—interest in learning, ability improvement, desire to learn behavior, and course satisfaction— showed significant differences (P<0.05) among students in the experimental group. In the self-assessment of English literacy, the dimensions with the most important proportion of “relatively large” ratings are learning motivation and class participation, accounting for 43.73% and 42.92%, respectively. The dimensions with the highest proportion of “substantial” grades are English learning ability and learning motivation, accounting for 21.03% and 20.63% in that order.