📁 About the project

Python project developed for System Integration Coding Challenge

The project consists in an implementation between a local Zammad Helpdesk with a Zammad API and two Python scripts. One for importing all tickets from a dataset of customer issues/complains huggingface and the other one for predicting the sentiment of the issue and correctly updating the priority of it using a model of sentimental classification.

📋 Run explanation and 💻 Requirements

Clone this repository with:

git clone https://github.com/masnik1/coding-challenge.git

At first you will need to run a local version of the Zammad ticketing tool running. Do that by going to (make sure docker is running)

cd zammad-docker-compose-master

And then :

docker-compose up

This will serve the application at http://localhost:8080/ - Then make sure to start it with default values. Maybe you'll need to create a new ZAMMAD_TOKEN Auth token and update the .env file. I suggest as an example selecting all preferences on Zammad Helpdesk (tried just admin and ticket api and it did not work -> Selecting all worked).

So go to:

1. Your profile photo on Zammad Helpdesk.
2. Profile.
3. Token Access.
4. Select all the permissions you need and copy the new token -> Update it on the .env file.

Later, we just need to set our env to run our python scripts: If you want to run it in a venv (to run the virtualenv venv command go to the project folder first):

pip install virtualenv
virtualenv venv

If you are running LINUX:

source venv/bin/activate

Windows:

venv\Scripts\activate

To install the required python libs just go to your project folder (or venv path) and run:

pip install -r requirements.txt

After that you can run first the: import_tickets_zammad.py and then the update_tickets_zammad.py

🚀 Citations

Transformer model used : https://huggingface.co/finiteautomata/bertweet-base-sentiment-analysis?text=I+like+you.+I+love+you Github repo : https://github.com/pysentimiento/pysentimiento