Execution of BERTopic topic modeling technique on Jira data to train the necessary models and extract topics from issues of every project.
The python requirements are available in file requirements.txt and may be installed using the command pip install -r requirements.txt. To execute BERTopic you must have set a MongoDB instance with the issues data. The necessary details must be set in file properties.py.
To execute the topic extraction, all properties in properties.py file must be set. After that, you can run 'python extract_topics.py' to perform the topic generation. You can insert the extracted topics for all projects to the MongoDB instance by running python add_topics_to_db.py.
After the topic extraction, by running python visualize_topics.py, a visualization of the topics extracted and the issues assigned to them can be produced for a selected project. The selected project must be set in the properties.py file.
If your use this tool or the corresponding dataset in your work, you can cite it using the following bibtex entry:
@unpublished{JiraApacheData,
author = {Themistoklis Diamantopoulos, Dimitrios-Nikitas Nastos and Andreas Symeonidis},
title = {Semantically-enriched Jira Issue Tracking Data},
year = {2023},
note = {Paper submitted}
}