Combining Textual Emotion Analysis and Gender Role Theory to Study the Portrayal of Women in Seven Male Maodun Literature Prize Winning Works

Abstract

In this paper, we construct a text sentiment analysis model based on BERT-Attention-BiLSTM, which is combined with the gender role theory to analyze the female character images in the award-winning works of male Mao Dun Literature Prize.The BERT-Attention-BiLSTM model combines the BERT model as a word vector training model, which can solve the word in word vectorization with the The BERT-Attention-BiLSTM model combines the advantages of the BERT model as a word vector training model in word vectorization, which can solve the problem of multiple meanings in word quantization, with the introduction of the Attention mechanism which distinguishes the degree of influence of different words in the text on the task of sentiment classification. By comparing with several text sentiment analysis models, the Bert-BiLSTM-Attention model has an accuracy of 81.20%, and the value of F1 is 58.42%, which is the best among all models. Then, the basic information and interpersonal network of seven Mao Dun Literature Prize works were quantitatively studied, and four works were selected for quantitative analysis of emotional trends, and the emotional trend lines portrayed fit with the plot development of the works. This study provides a literary perspective reference for understanding the interaction between female development and social progress.