Design of a literature-based poetry appreciation system based on perceptron model

With the continuous promotion and popularity of the digitalization of big data information, it has gradually penetrated various industries and played an increasingly important role in people’s lives. In this paper, the design of a literary poetry appreciation system is studied using a perceptron model, focusing on the construction of a literary poetry appreciation system through a generalized multi-layer perceptron model structure and the evaluation of metrics on the poetry appreciation system dataset using the GBP training algorithm and genetic training method. In the generalized multilevel perceptron model, the data indicators are classified into four categories: poetry music, poetry sound effects, poetry vocals, and other types, and the accuracy rates of the four categories of indicators are analyzed. The analysis shows that the accuracy rate of all four categories of indicators is between 92% and 97%, indicating the feasibility of the poetry appreciation system. The generalized multilevel perceptron model was used to analyze and evaluate the students from the ability perspective. It can be seen that the students’ data showed only a small and insignificant improvement in the dimensions of literacy, comprehension, and analysis, while the scores in the dimensions of reasoning, strategy, and expression improved significantly. It provides a good research perspective for perceptual machine model-assisted teaching and cultivating students’ quality from a competence perspective, which is of historical importance for developing and transmitting Chinese literature.