Overview

Creating medical chatbot using Python (Flask) and machine learning algorithms for purposes of project for the course of Intelligent Systems at the Faculty of Organization and Informatics, University of Zagreb.

Requirements

Python 3.5 or newer.

Dependencies:

• Flask
• PyTorch
• NLTK
• NumPy
• Scikit-learn
• Pandas
• matplotlib

Install requirements

Before running the application you need to install the dependencies. We recommend to use the virtual environment virtualenv for this.

Linux:

python3 -m venv venv
venv/bin/activate
pip install flask torch nltk numpy sklearn pandas matplotlib

Windows:

py -3 -m venv venv
venv\Scripts\activate
pip install flask torch nltk numpy==1.19.3 sklearn pandas matplotlib

In order for nltk tokenization to work, the 'punkt' package must be downloaded. To do this, simply enter the Python shell and run the following:

import nltk
nltk.download('punkt')

This will install all the required dependencies needed to run the application successfully.

Run

To run MedicalChatbot, cd into MedicalChatbot repo on your computer and run python -m flask run. This will run the Flask server in development mode on localhost, port 5000.

* Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)