On the Impact of Requirements Smells in Prompts: The Case of Automated Traceability

<pre>This is the dataset for the paper <i>On the Impact of Requirements Smells in Prompts: The Case of Automated Traceability</i>. It contains the following data:<br><br><i>evaluation.ipynb</i>: The Jupyter notebook containing our evaluation.<br><i>games/</i>: The code and requirements in CSV format for all games.<br><i>ground_truth.csv</i>: The ground truth in CSV format.<br><i>llm-request-code/</i>: The code used to generate the samples and prompt the LLMs via APIs.<br><i>prompts/</i>: The prompts used with the LLMs.<br><i>requirements_and_ground_truth.xlxs</i>: The requirements and ground truth (for both raters) as an Excel spreadsheet.<br><i>results_all.json</i>: The results retrieved from the LLMs in JSON format.<br><i>smell_types_and_categories.pdf</i>: A PDF file listing all considered smell types with their categories.<br><br></pre>