Research on Emotional Classification and Literary Narrative Visualization Based on Graph Convolutional Neural Network.

Watching a movie for three minutes has become a popular term in contemporary life, and people also hope to spend less time understanding the emotional direction and general plot of long novels. Currently, research on emotional analysis in novels mainly focuses on foreign languages. This article introduces Graph Convolutional Neural networks into the text emotion classification of novels. Text GCN is used to conduct emotion analysis on Ba Jin’s novel “The Rapids Trilogy”, and iStoryline is used to visualize the storyline of literary works. Visual coding adjustments are made based on the results of emotion analysis in the generated storyline, where line aggregation and dispersion represent whether the characters appear in the same scene, The brighter the color of the lines, the more positive the emotions, making the visualization more vivid, full, and beautiful. The experimental results indicate that this method enables readers to easily grasp the overall direction of the novel’s plot and have a clearer understanding of the emotional changes of the characters. In addition, the accuracy of the Text GCN model is superior to traditional sentiment classification methods on Chinese datasets. This method provides a new approach for emotional analysis and narrative visualization research in literary works, and can be extended to other fields of digital entertainment.