MARCUS: An Event-Centric NLP Pipeline that generates Character Arcs from Narratives

2025-10-20

AI Summary

This study addresses the challenge of quantitatively modeling character development in narrative texts. We propose the first computational, event-centric, relation-aware dynamic arc model that formalizes the literary concept of the “character arc” as a traceable and quantifiable temporal graph. Methodologically, we design an end-to-end NLP pipeline that jointly extracts narrative events, identifies character participation, infers implicit sentiment polarity, and aggregates inter-character interactions across paragraphs. Experiments on long-form series—*Harry Potter* and *The Lord of the Rings*—demonstrate the model’s effectiveness in capturing multi-character sentiment evolution and shifting relational tension, yielding highly interpretable arc visualizations. Our primary contribution is the first automated, relational, and event-driven characterization of narrative arcs, establishing a novel paradigm for computational narratology and digital humanities.

Technology Category

• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Reasoning under Uncertainty: Relational Probabilistic Models
• Cognitive Modeling & Cognitive Systems: Computational Creativity

Application Category

• Web Mining and Content Analysis: Models for Web evolution
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Character arcs are important theoretical devices employed in literary studies to understand character journeys, identify tropes across literary genres, and establish similarities between narratives. This work addresses the novel task of computationally generating event-centric, relation-based character arcs from narratives. Providing a quantitative representation for arcs brings tangibility to a theoretical concept and paves the way for subsequent applications. We present MARCUS (Modelling Arcs for Understanding Stories), an NLP pipeline that extracts events, participant characters, implied emotion, and sentiment to model inter-character relations. MARCUS tracks and aggregates these relations across the narrative to generate character arcs as graphical plots. We generate character arcs from two extended fantasy series, Harry Potter and Lord of the Rings. We evaluate our approach before outlining existing challenges, suggesting applications of our pipeline, and discussing future work.

Problem

• Computationally generate event-centric character arcs from narratives
• Extract events, characters, emotions to model inter-character relations
• Track narrative relations to create graphical character arc plots

Innovation

• Extracts events, characters, emotions, and sentiment
• Models inter-character relations across narratives
• Generates character arcs as graphical plots