An Optimization Study of Artificial Intelligence in Teaching Chinese as a Foreign Language for Present and Contemporary Literature Lecture

Abstract

This paper takes artificial intelligence as the starting point of teaching optimization, firstly converts the present and contemporary literature into feature vectors that can be used for classification, and extracts word frequency spatial features according to the keywords. Then the lexical annotation is carried out to complete the participle extraction, and the word frequency weights are counted in order to explain the present and contemporary literature in a targeted way. Next, a convolutional neural network classification model is constructed to enhance the representation of current and contemporary literature by adding input information so that the model learns richer features. Integrate the historical learning data with the learner's forgetting mechanism to solve the problem of learning learner's knowledge forgetting. Finally, the corresponding weights of the historical data are calculated by utilizing the self-attention mechanism and the forgetting mechanism, according to which the optimal form of explanation is generated. Finally, empirical validation shows that the feature word visualization similarities are all higher than 0.6, and the classification error rate of the convolutional neural network is only 1.16% compared with other algorithms. In the teaching effect, the highest average score of the oral class grade after optimization is 93.65, and the highest student participation in the explanation of modern and contemporary literature is 99.92%. Artificial intelligence enriches and changes the traditional means of teaching Chinese as a foreign language, improves the teaching quality and optimizes the teaching effect