juanesc396 / newspaper-analytics-webapp

This is the last part of the project that analizes newspapers around the world

Newspaper Analytics: Web App

In this repo, I will save the web app in where I show the data recolected and analized by the scraper and NLP model.

Technical Details

The technologies used to build the project are:

• Dash
• Plotly
• Pandas
• Nginx
• Gunicorn
• AWS RDS -> MySQL
• EC2 -> Web Hosting

Design

In my opinion, newspapers should represent sincerity and transparency, for that reason I chose a glassed style and colors that represent those emotions.
I used css to styling the web.
When I chose the fonts, I quickly leaned towards Serif fonts, but they didn't meet my expectations. That's why I chose a Sans Serif font that fits with the web style.

For the Future

I want to:

• Improve the loading times. In the future I will try to implement Polars, a library write in Rust, that allows data process faster.
• Improve the design.
• Add graphs like Scatter Plot, to show how the positivity varies over time.
• Use a better NLP model to get more accurated predictions.