Retain or Reframe? A Computational Framework for the Analysis of Framing in News Articles and Reader Comments

2025-07-06

AI Summary

Prior NLP approaches detect framing in news articles or reader comments in isolation, neglecting the dynamic co-construction between media agendas and public responses. Method: We propose the first large-scale computational framework for cross-textual framing analysis, jointly modeling source news content and audience reactions. Our hierarchical architecture first performs sentence-level frame prediction, then reconstructs document-level dominant frames; it incorporates a topic-matching–driven article-comment alignment mechanism, a fine-grained frame taxonomy, a multi-label Transformer classifier, and a frame aggregation algorithm. Contribution/Results: Evaluated across 11 sociopolitical issues and two major news outlets, our framework demonstrates both widespread frame reuse and issue-specific framing patterns. We release a high-performance cross-domain frame classifier, a manually annotated dataset, and a large-scale frame prediction dataset—establishing a foundational resource for computational framing research.

Technology Category

• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining
• Application Domains: Humanities & Computational Social Science
• Knowledge Representation and Reasoning: Argumentation

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Social Networks and Social Media: Computational social science
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

When a news article describes immigration as an "economic burden" or a "humanitarian crisis," it selectively emphasizes certain aspects of the issue. Although extit{framing} shapes how the public interprets such issues, audiences do not absorb frames passively but actively reorganize the presented information. While this relationship between source content and audience response is well-documented in the social sciences, NLP approaches often ignore it, detecting frames in articles and responses in isolation. We present the first computational framework for large-scale analysis of framing across source content (news articles) and audience responses (reader comments). Methodologically, we refine frame labels and develop a framework that reconstructs dominant frames in articles and comments from sentence-level predictions, and aligns articles with topically relevant comments. Applying our framework across eleven topics and two news outlets, we find that frame reuse in comments correlates highly across outlets, while topic-specific patterns vary. We release a frame classifier that performs well on both articles and comments, a dataset of article and comment sentences manually labeled for frames, and a large-scale dataset of articles and comments with predicted frame labels.

Problem

• Analyzes framing in news articles and reader comments
• Develops computational framework for frame alignment
• Examines frame reuse patterns across topics and outlets

Innovation

• Computational framework for framing analysis
• Sentence-level frame prediction and alignment
• Frame classifier for articles and comments