This Flask application uses Gensim's LDA module to model the top words of the two most important topics of all the Yelp reviews for a selected restaurant.
These top words are displayed as two word clouds, which allows a potentional user to gain impressions of a chosen restaurant without having to read any of the actual reviews!
The application can be run locally by setting up a python virtual environment in the same directory as the code and then installing the requirements from requirements.txt:
python -m venv env
source evn/bin/activate
pip install -r requirements.txt
python application.py
That's it! The application is now available on your local browser to play with and change to your heart's content.
You can even install the application - as is - on AWS Elastic Beanstalk, by compressing all the files with a zip program and then uploading the zipped file to a fresh Elastic Beanstalk application using your own AWS account.
Due to the English language model included for Spacy, it could get expensive, however. ;)