Sports, Politics, and Social Media: A Human-AI Collaborative Analysis of Consumer Reactions to Trump’s Break 50 Appearance

Thanks for your interest! This repository contains the code, prompts, fine-tuning data, and outputs associated with the paper "Sports, Politics, and Social Media: A Human-AI Collaborative Analysis of Consumer Reactions to Trump’s Break 50 Appearance." The study examined audience discourse surrounding Bryson DeChambeau and Donald Trump’s Break 50 video using a human-AI collaborative framework.

Repository Structure

Folder/File	Description
BERTopic/BERTopic_hierarchical_clustering	Hierarchical clustering outputs from BERTopic modeling.
BERTopic/BERTopic_topic_results	Topic modeling results, including topic distributions and keywords.
Fine-tuning/prompt_gpt-4.1	Prompt used with GPT-4.1 for generating fine-tuning data.
Fine-tuning/ft_gpt-4o-mini_training	Training dataset used for fine-tuning GPT-4o-mini.
Fine-tuning/ft_gpt-4o-mini_validation	Validation dataset used for fine-tuning GPT-4o-mini.
Reasoning/prompt_o4-mini	Prompt used with o4-mini for reasoning runs.
Reasoning/output_o4-mini	Outputs generated using the o4-mini model in reasoning runs.
Appendix	Supplementary files provided as an appendix to support verification and replication.

Usage

1. Clone the repository and navigate into the desired folder.
2. Explore the BERTopic folder for topic modeling outputs and hierarchical clustering files.
3. Use the Fine-tuning folder if you wish to replicate or adapt the fine-tuning of GPT-4o-mini with the provided training and validation sets.
4. Refer to the Reasoning folder for prompt templates and example outputs from reasoning experiments.
5. Supplementary details and additional resources are available in the Appendix.

These resources are designed to provide transparency, replicability, and adaptability for future research at the intersection of sport, politics, and AI-assisted analysis.

Citation

If you use this repository or build upon this work, please cite the associated paper:

@article{qian2025break50,
  title={Sports, Politics, and Social Media: A Human-AI Collaborative Analysis of Consumer Reactions to Trump’s Break 50 Appearance},
  author={Qian, Tyreal Yzihou and Gong, Hua and Xu, Chenglong},
  journal={Journal of Retailing and Consumer Services},
  year={2026},
}

Contact

For questions, suggestions, or collaboration opportunities, please contact: Tyreal Yizhou Qian.