We wanted to use Markov Chains to create a word generator of some sort, and decided that creating a news aggregator would be a cool and expandable project. Many features felt quite separate from the last, so new ones can quickly be added with little risk of breaking anything or not finishing.
What it does
Users can search by keywords that connect directly to the NewsAPI. Additionally, they can generate Markov Chain based article names and descriptions for fun.
How we built it
We used Python, Flask, HTML, Bootstrap, and CSS.
Challenges we ran into
None of us had too much experience using Flask, and deciding how to connect the frontend to the backend took a while. Additionally, implementing the article generator was quite difficult as the algorithm was a bit nuanced.
Accomplishments that we're proud of
It felt great to finish everything we planned on the night before; we were worried that would not be possible and we would be on a time crunch. After this, we continued to push to add new features such as sorting by topic/genre, a back button, and some extra visual additions such as the icon in the tab bar.
What we learned
We learned how to create simple full stack applications. Although we could have used a more nuanced front-end system, if we had more time we would likely change to a more fully-fledged framework such as Django instead, anyways.
What's next for NewzFlash
We could add citation creators, more animations, a better search engine, searching by country, more sorting, and more. With a business subscription to NewsAPI, we could also have many times more search results; if we had access to more results, we could reasonably added result pages instead of having everything come out at once. Additionally, we could turn the site into somewhat of a social media application by adding accounts and features like commenting.