Python module for explore interactional and customer journey data. A little bit of natural language processing (NLP), machine leaning (ML), and large language models (LLMs) to help you better understand what people are saying or writing.
This project requires Python version range >=3.9 to <3.12. Below are the instructions to set up the environment for this project.
You can install the interactovery module from PyPI:
pip install interactoveryAfter installing the required packages, you need to download the spaCy English model "en_core_web_lg". Run the following command:
python -m spacy download en_core_web_lgFor Windows users, to install hdbscan, you may need to install the build tools from Visual Studio. This is because hdbscan requires C++ compilation which is not natively supported in Windows Python environments.
You can download the Visual Studio Build Tools from this link. Follow the instructions to install the necessary components.
There is a GitHub project with various Jupyter notebooks to explore usage.
This module is something I am putting together to allow everyone to have easy-to-use tools to analyze and understand their interactional data. This project is not polished production code that has been battle hardened, but something I am building publicly as part of a blog series that will improve over time. Sometimes, those changes are going to be breaking. I'll aim for all the best practices around backwards compatibility once we hit v1.0.0.
Happy coding! :)