Answers are grounded in this paper and cite their sources.
This study explores how Donald Trump’s political persona is constructed through media framing in the British newspaper The Guardian using Natural Language Processing (NLP) methods. Employing a mixed-methods approach, the research combines computational text analysis with qualitative frame interpretation to examine both large-scale patterns and discursive features of news coverage. The study situates media framing analysis within the broader context of digital political communication and contemporary journalistic practices. The dataset consists of 47 articles published between July 2024 and September 2025, addressing major political and international events. Following standardized text preprocessing, topic modeling, sentiment analysis, named entity recognition, and trend detection were applied. The analysis revealed four dominant and recurring frames (foreign policy, leadership, domestic policy, and conflict), with conflict and foreign policy predominating across the corpus. Sentiment analysis showed a predominantly negative tone, with negative sentiment comprising approximately 51–64% of the coverage. An additional LLM-based procedure enabled the identification of key frame components and narrative roles shaping Trump’s representation. This approach allows for the systematic examination of both explicit evaluations and implicit narrative structures embedded in political news discourse. The findings demonstrate that NLP-based methods provide effective tools for detecting evaluative and ideological patterns in political media discourse. By integrating quantitative techniques with theory-driven qualitative analysis, the study offers a nuanced account of how linguistic and narrative strategies contribute to the construction of political personas in contemporary digital journalism.
No takes yet. Share an insight, caveat, or question.
Micallef et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: