The Application of Intelligent Evaluation Method with Deep Learning in Calligraphy Teaching

Scientific and effective teaching quality evaluation (QE) is helpful to improve mode quality. At present, calligraphy (CT) QE methods are few in number have poor effect. Aiming at these problems, deep learning (DL) introduced realize intelligent of CT First, based on relevant research, the CTQE indicator system constructed. Secondly, rough set principal component analysis (PCA) used reduce dimension index extract four common factors. Then, corresponding data input into BP neural network (BPNN) model optimized by improved sparrow search algorithm for fitting. Finally, combining above contents, (ISSA) BPNN built The experimental results show that loss value ISSA-BPN 0.21, fitting degree 0.953. Accuracy 95%, Precision 0.945, Recall 0.923, F1 0.942, AUC 0.967. These values superior most advanced available. SSA-BPNNCTQE proposed study has excellent performance CTQE. This positive significance improvement students' level.