Reading through the NewsQ challenge, we found the real-world applications of the project really interesting. Because our team had members with interests that branched out from computer science, the idea of creating back-end code

What it does

It takes into consideration a variety of factors which are assigned weights based off their relative importance. The weighted values for each factor are used to calculate a final rating for an article. The articles are then arranged according to the ratings.

How I built it

We used Python as the framework for working with different API's to pull data

Challenges I ran into

Collecting website data, using effective APIs for our algorithms, debugging for various errors when utilizing API and dealing with data

Accomplishments that I'm proud of

When we were almost done with our project, an API stopped working and we felt doomed for a short time. When we were able to work through these issues, there was an amazing sense of accomplishment that came over the team, and it lifted our spirits. Accomplishments like this are a large aspect of what makes hackatons a good time!

What I learned

We learned data scraping the web with multiple APIs, developing algorithms to quantify that scraped data, and outputting that data in a spreadsheet

What's next for Optimizing News Recommendation Algorithms

  • Finetune the algorithm and add more factors to make the article ranking more accurate (ex: check amount of ad content, utilizing a UK fact checker, etc.)

  • Utilize machine learning by creating a training set and applying attributes known to make an article more or less reliable

  • Create a front end interactive user-friendly interface to allow users to browse unbiased and trustworthy articles of their interest

Built With

  • media-cloud-api
  • newspaper3k
  • python
  • sentiment-analysis-api
Share this project: