Analyzing the Characteristics of Female Emotional Expressions in the Works of Chinese Filipino Women Writers Based on Emotional Computing Models

Abstract

Applying sentiment recognition and computation to literary works has gradually become a popular direction in recent digital humanities research. The study introduces the sentiment computation model into the analysis of Filipino Chinese female writers' works, effectively extends the basic sentiment lexicon through the improved SO-PMI algorithm, constructs the sentiment computation model, and carries out experimental analyses and empirical studies on the sentiment analysis of Filipino Chinese female writers' works. The experimental analysis verifies the advantages of the proposed sentiment computation model in the task of text sentiment analysis as well as the effectiveness of the sentiment lexicon, and the F1 values of the improved SO-PMI algorithm for the classification of positive, neutral, and negative sentiments are improved by 11.96%, 6.96%, and 6.89% over the SO-PMI algorithm. The percentage of emotions in the works of Filipino Chinese women writers is dominated by sadness, fear, and anger, which all account for more than 16%, reflecting the difficult situation, identity entanglement and inner sorrow of Filipino expatriates. The emotion calculation model can reveal and summarize the structure of emotion in narrative texts, providing a richer and more three-dimensional perspective for Chinese literary criticism.