Focal Interest
Focal Interest
AI Summary of Peer-Reviewed Research

This page presents an AI-generated summary of a published research paper. The original authors did not write or review this article. [See full disclosure ↓]

Publishing process signals: STRONG — reflects the venue and review process. — venue and review process.

Transformer method measures inter-party communication

A man in a dark suit speaks at a podium with a blue EU flag beside him while four other professionals seated at a table listen during a formal political or governmental meeting in a modern conference room.
Research area:Social SciencesPolitical Science and International RelationsComputational and Text Analysis Methods

What the study found

The study presents a transformer-based approach for measuring inter-party communication, defined here as public communication by parties about other parties with a positive, neutral, or negative stance on collaboration, policy, or personal issues. The authors report that the approach can automatically classify large volumes of text.

Why the authors say this matters

The authors say the study helps deepen understanding of party competition, advances methods for automated text classification, and enables new research on political communication. They also note that research on inter-party communication has been limited by a lack of conceptual clarity and by difficulty measuring it at scale.

What the researchers tested

The researchers introduced a new transformer-based text-classification method and tested it in case studies on coalition signals in Germany and negative campaigning in Austria. The abstract describes the method as designed to automatically classify large volumes of text.

What worked and what didn't

The abstract says the case studies demonstrated the approach's effectiveness. It does not describe any failures, comparative weaknesses, or detailed performance results.

What to keep in mind

The available summary does not provide quantitative results, validation details, or specific limitations. It also does not describe how the model performed across different kinds of inter-party communication beyond the two case studies.

Key points

  • The article defines inter-party communication as public communication by parties about other parties, with positive, neutral, or negative stance.
  • It introduces a transformer-based approach to automatically classify large volumes of text.
  • Case studies covered coalition signals in Germany and negative campaigning in Austria.
  • The abstract says the approach was effective in those case studies.
  • The authors say the study may support new research on political communication and automated text classification.

Disclosure

Research title:
Transformer method measures inter-party communication
Authors:
Anna Adendorf, Oke Bahnsen, Thomas Gschwend, Lena Maria Huber, Simone Paolo Ponzetto, Ines Rehbein, Lukas F. Stoetzer
Institutions:
University of Mannheim, Witten/Herdecke University
Publication date:
2026-02-24
DOI:
10.1017/psrm.2026.10094
OpenAlex record:
View
AI provenance: This post was generated by OpenAI. The original authors did not write or review this post.

More posts