News 451

Inspiration

The news is an important form of connection for the world, allowing people to know what is happening from the other side of the world. However, fake news is a huge problem not only in America but in the world alike. It is yellow journalism/propaganda that consists of deliberate disinformation primarily spread through broadcast news media or online social media. Often times, it is difficult for people to know what news sources or articles are reliable, and which ones are not. Additionally, people tend to only read sensational news articles that are not usually accurate. We sought to solve this issue by providing a novel application helps people know what news articles they can trust. This will allow everyone to better understand the news and allow for more educated and informed opinions regarding political matters.

What it does

How we built it

To build the application, we used javascript, jsx, and css3 languages. We also used React javascript library for building user interfaces and the material UI framework. For the machine learning neural network, we used Scikit learn and Python programming language. We built the app by having popular news articles obtained by scraping various news sources. We wanted our users to not have to worry about obnoxious paywalls and ads that many companies put in place so we scrape the article for the article content and relevant images. We also pressed our keyboards a lot.

Challenges we ran into

We ran into many challenges throughout the programming experience that we overcame. One such issue was that we had difficulty in making a user interface that was sufficiently simple that easily navigable for users. The main source of problems came from making the machine learning model. Figuring out the correct parameters for the network took a lot of time.

Accomplishments that we're proud of

We are proud of many things in this application. We are proud of a working machine learning neural network that can successfully determine the reliability of a news source or article 92.3% of the time. We are proud of the intuitive and simple user interface we created for the application, allowing users to easily use the app and successfully filter out unreliable news and better understand the news.

What we learned

We learned how to make a simplistic user interface that is easy to use. We also learned how to preprocess a large dataset that gets fed into a Machine Learning Algorithm. This was our first project using multiple programming languages so we had to learn how to integrate them together.

What's next for News 451

In the future, we are going to add more features to the application, as well as improve existing elements of the app. We are planning on adding better navigation of the application through different buttons and pages that allow users to view popular news in specific categories ie. sports, technology. We are also planning on improving the machine learning model to corroborate with other news sources to make it even more accurate and provide a better estimation of the reliability of articles, striving for 100% accuracy.