Construction of teaching quality evaluation model of online dance teaching course based on improved PSO-BPNN

Abstract With the combination of Internet technology and teaching methods, the online dance teaching mode is becoming more popular. However, traditional teaching quality evaluation methods such as neural network models have low efficiency, insufficient accuracy, and are prone to ignoring individual differences, making it difficult to better adapt to the online dance teaching mode. Accordingly, this study constructs a quality evaluation index system for online dance teaching by consulting relevant online course teaching experts, considering individual student differences, and existing research, including preparation, classroom teaching implementation, classroom teaching effectiveness, and reflection on after-school teaching. On this basis, the network dance teaching quality evaluation model based on back-propagation neural network is constructed. For these shortcomings, it is optimized by the improved particle swarm. This study uses inertia weight and crossover operator strategy to optimize particle swarm optimization algorithm, achieving the collaborative work of back-propagation neural network and particle swarm optimization. This study uses statistical data from online dance teaching courses in universities for experimental analysis. According to the findings, the accuracy of the designed model reached 97.25%, which significantly exceeded other commonly used methods. Based on this evaluation model, both teachers and students can receive feedback on teaching and learning processes, thereby improving teaching methods, meeting personalized needs of students, and enhancing learning outcomes. This study combines PSO and BPNN to provide a new method for evaluating the quality of online dance teaching, providing new reference ideas for improving the quality of online education in the information society.