MyCI - My Campus Information Chatbot

This is a Telegram-bot which is designed and developed as part of the CS4145: Crowd Computing course at the Delft University of Technology by team CrowdIT - Neha Sree Thuraka, Sharad Shriram, Salim Salmi, Yizi Chen. The use case for which the chatbot is developed is course recommendation using the concept of crowd computing. This is prototype is based on the following flowchart,

Using the chatbot

The chatbot is accessible at . However, ping us on the course slack channel for you to interact with the chatbot. Once you are inside the chatbot, you can use the following commands to interact with it

- `/start` starts the interaction with the chatbot
- `/display` displays the course preferences of the user
- `/recommend` gives recommendations of courses based on user preferences
- `/addpreference` updates the course preferences for the user.
- `/points` displays the points accumulated 
- `/remove` remove the preference of the user
- `/menu` lists all the chatbot commands

Project Structure

The project directories and files are structured in the following tree:

- `images` images for the repo
- `python-bot` 
    - `answer.py` processing and aggregating answers recieved from tasks
    - `bot.py`    handles bot instances for handling messages and updates
    - `courselist.csv` compiled list of course title and course codes
    - `main.py` starts the bot, maps the commands to respective functions
    - `mongodb.py` to handle MongoDB to create and modify data relating to the task and user model
    - `neha.py` intial implementation of task allocation done by [@neha](https://github.com/nehasreet)
    - `question.py` module to get questions/ queries from user
    - `recommendations.py` provides course recommendations based on `csv` file
    - `requirements.txt` third-party modules that are required
    - `settings.py` set the access code and timeout period of request for the bot
    - `taskallocation.py` create and allocate tasks to similar users 
    - `usermodel.py` managing the user model based on the chat messages
- `README.md`

Setting Up

Before getting started you'll need the following software installed on your operating system:

• Python 3
• MongoDB (If you're using Windows, this link is useful)

The following python-packages are the next that needs to be installed. Don't forget to be using Powershell/ Terminal with Administrator or super-user privileges. To make your lives easy you can paste the following command to install the packages using pip. Do not have pip? Check this out:

pip install -r requirements.txt

Local Testing

If all the installation went well, you can launch Powershell/ Terminal and run the following command to start the bot service:

python main.py

And in Telegram search for MyCI_bot and you're good to go! Have fun!

In the event of any trouble create an issue on this repo. We'll address it as soon as possible.