Leveraging Human Expertise for LLM-Assisted Dialogue Character Extraction and Attribution in Classic Chinese Novel

Classic Chinese Novels - Using Distance Reading to Understand Complex Narratives

· Revealing new insights into character relationships and narrative roles...

Social Network Analysis - Requiring Accurate Data Extracted from the Novels

· Challenges in classic Chinese novels

· Challenges in data extraction

• Complex character networks
• Rich contexts
• Multiple aliases per character

Dream of the Red Chamber

• Manual extraction of characters from classic Chinese novels requires significant human effort and is highly time-consuming.
• The richness of context in novels and the complexity of character references present challenges for fully automated extraction methods.

Text Upload

Original novel text upload

Rule-based Segmentation

Break into dialogue units with context

Speaker and Listener Extraction

prompt = f"\nQ: I will give you a dialogue sentence and a passage of context. Please repeat the dialogue sentence, then based on the context, identify the speaker, the primary listener(s), and the secondary listener(s) in the specified dialogue sentence. Dialogue sentence: {talk}. Context: {context}. Please provide your answer in the format: 'Dialogue sentence: [dialogue], Speaker: [speaker], Primary Listener(s): [primary listener(s)], Secondary Listener(s): [secondary listener(s)]'. Note: 1. Use commas (",") to separate the dialogue sentence, speaker, primary listener(s), and secondary listener(s); 2. If there are multiple speakers or multiple primary/secondary listeners, separate them using "、"; 3. Do not insert any line breaks in your answer; 4. Resolve pronoun references carefully and avoid vague references like "you", "I", "he", etc.; 5. Do not include any explanation or analysis in your answer — treat it like a fill-in-the-blank question. The more concise, the better; 6. If the speaker, primary listener(s), or secondary listener(s) cannot be identified, respond with 'None'. \nA:"

⇢ Add to Database

Resolve Co-references

prompt = f"\nQ:This is a fill-in-the-blank question with an answer of 0 or 1. Please determine: In the book '{chinese_book_name}', are '{entity}' and '{main_entity}' the same character? If yes, return 1; if not, return 0. Do not consider literary implications. No explanation or analysis is needed." \nA:"

⇢ Add to Database

Data visualization design

• Large nodes (green) = main name
• Small nodes (yellow) = aliases
• Links between large nodes = relations (measured with numbers of conversation)
• Chapter-specific networks: generating dialogue networks for all character relationships up to the current chapter, revealing narrative progression.

Manual Disambiguation

Correct extraction errors

Key Characters

Important intermediary characters identified

Interaction Patterns

Social dynamics revealed

AI + Human

Improving Algorithm Efficiency

Improve the speed of data processing, thus generate the result faster.

Incorporating SNA Algorithm

Use social network analysis algorithm to build a final network, which can directly serve for the literary analysis.