Who Attacks, and Why? Using LLMs to Identify Negative Campaigning in 18M Tweets across 19 Countries

2025-07-23

AI Summary

This study addresses the challenge of efficiently and scalably identifying negative campaigning behaviors in transnational political competition, particularly within massive, multilingual social media text corpora. Methodologically, it introduces the first zero-shot large language model (LLM)-based framework for cross-lingual automatic classification of negative political content—eliminating reliance on manual coding or supervised learning—and validates its high accuracy and reproducibility on a multilingual benchmark dataset. Applying this framework to 18 million tweets across multiple countries, the analysis reveals systematic associations between party type and adversarial rhetoric: governing parties exhibit significantly fewer negative expressions, whereas ideologically extreme and populist parties—especially far-right ones—deploy markedly more aggressive language. The core contribution is the development of the first zero-shot, cross-lingual, and scalable analytical paradigm for political communication research, establishing a novel methodological foundation for large-scale comparative studies of political discourse.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Machine Learning: Large Multimodal Models (LMMs)
• Game Theory and Economic Paradigms: Adversarial Learning

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Negative campaigning is a central feature of political competition, yet empirical research has been limited by the high cost and limited scalability of existing classification methods. This study makes two key contributions. First, it introduces zero-shot Large Language Models (LLMs) as a novel approach for cross-lingual classification of negative campaigning. Using benchmark datasets in ten languages, we demonstrate that LLMs achieve performance on par with native-speaking human coders and outperform conventional supervised machine learning approaches. Second, we leverage this novel method to conduct the largest cross-national study of negative campaigning to date, analyzing 18 million tweets posted by parliamentarians in 19 European countries between 2017 and 2022. The results reveal consistent cross-national patterns: governing parties are less likely to use negative messaging, while ideologically extreme and populist parties -- particularly those on the radical right -- engage in significantly higher levels of negativity. These findings advance our understanding of how party-level characteristics shape strategic communication in multiparty systems. More broadly, the study demonstrates the potential of LLMs to enable scalable, transparent, and replicable research in political communication across linguistic and cultural contexts.

Problem

• Classifying negative campaigning in multilingual tweets efficiently
• Comparing LLM performance with human coders and traditional methods
• Analyzing cross-national patterns of negativity in political communication

Innovation

• Uses zero-shot LLMs for cross-lingual classification
• Analyzes 18M tweets across 19 European countries
• Outperforms human coders and supervised machine learning