This repository hosts the code and resources for a project focused on classifying user stories based on their susceptibility to privacy disclosure. It utilizes FlairNLP for effective text classification.
- src: This folder contains the source code of the project. Here, you'll find all the scripts and modules required to run and train the classification models.
- data: The 'data' directory is organized into 5 different folds for cross-validation purposes. This setup facilitates the evaluation of the model's performance across various unseen datasets.
- flair-results: All raw results obtained from running the classification models using FlairNLP are stored here. This includes output logs, intermediate results, and any error reports generated during the execution.
- performance-results: For a cleaner and more comprehensive view of the results, we have processed and compiled them into neat Excel files in this folder. These files allow for easier analysis and comparison of the model's performance across different folds and parameters.
Coming soon