Exploring the structure of micro topics in Chinese language and literature based on the fusion of multiple features

In this paper, firstly, the topic structure of the Chinese language and literature is explained, and the micro topic structure representation system is constructed by using the main statement bit structure theory and the main bit advancement model. Secondly, the word2vec network is used to extract the features of Chinese language and literature elements, and a multi-feature fusion classification recognition model is constructed based on SVM. Finally, the corpus was used to validate the example. The results show that the percentage of radial primary advancement pattern of micro topic structure is the highest and reaches 94.13%, and the recognition accuracy of SVM-based multi-feature fusion recognition classification model for micro topic structure is improved by 12.04%, 18.29% and 7.45% respectively compared with BTM algorithm, LDA algorithm and K-means algorithm.