Paper: Uncovering the Causes of Emotions in Software Developer Communication Using Zero-shot LLMs

In the following, we briefly describe the components included in this project and the software required to run the experiments.

Project Structure

The project includes the following files and folders:

• /annotation: A folder that contains annotation instructions and the annotated file.
• /blue_score: A folder that contains the scripts for BLEU score computation for emotion-cause extraction.
• /cluster: A folder that contains the scripts for generating clusters for the case study.
• /Emotion_Classification: A folder that contains scripts for training and testing emotion classification using BERT and RoBERTa.
  • emotion_classification.py: the script of emotion classification using fine-tuned LLMs.
  • esem-e.py: the script to run eseme mode.
  • ReadMe.md: contains the readMe on how to run emotion_classification.py and esem-e.py.
  • /Dataset: A folder containing the experiment's datasets.
• /Prompts: A folder that contains prompts that have been used in this experiment.