Hierarchical Analysis of the Relationships between Main Characters in the First 80 and the Last 40 Chapters of a Dream of Red Mansions based on Natural Language Modelling

Natural language processing is important in computers and big data. Preliminary analysis and organization of text through natural language processing is one of the key research contents. The traditional character relationship in classical masterpieces is one of the main means of computer interpretation of literature as it allows readers to understand the character relationship and storyline clearly. We analyzed the literary works using natural language processing and extracted the character relationships through text analysis. The main characters of ‘Dream of Red Mansions--Jia Baoyu and the Twelve Haircuts of the Golden Hairpin’ were used as the object of analysis. Using the social network analysis, the relationship levels of the main characters were explored in the first eighty and forty times. The relationship map of the main characters in the work was drawn. The characters corresponded to the nodes on the social network, and the relationships between the characters corresponded to the edges of the social network. By analyzing the text, the character relationship network of the first eighty and the last forty chapters was obtained. The results showed that the relationship levels of the main characters were different, indicating that the character relationships hanged significantly.