Inspiration

In the current COVID world, social media is an important way for us to connect with friends and family. Knowing this, we wanted to create something that enables users to see a version of their own digital footprint.

What it does

Uses Facebook's API and machine learning to generate aesthetic word cloud based on user's Facebook posts.

How we built it

For the frontend, we used React in conjunction with HTML, CSS, and JavaScript. For the backend, we used JavaScript as well; we explored using APIs by utilizing FaceBook's Graph API to scrape the users posts' data and process the data. Then, we used the WordCloud machine learning library to generate awesome visuals, with the words in the visual sized based on how frequently words of that topic appeared in the user's FaceBook posts!

Challenges we ran into

The tight deadline for the project and different time zones for literally every member of the group made communication and GitHub merging difficult. With limited time, we also faced the challenge of quickly learning and applying relevant API to our project. None of us had used FaceBook's API before and learning to grab data was difficult, but rewarding.

Accomplishments that we are proud of

Every coin has two sides; with the unfortunately rushed experience of the hackathon, we gained new friendships and showed resilience and team spirits in face of the ticking clock. Knowing that we started late, we quickly concentrated our plan and got right down to business. In the end, we are able to create a we

What we learned

We have learned to use the Facebook API to request a user's posts' information and then use machine learning to generate a word cloud. It has been a valuable experience during which we realized the importance of and learned incorporating a variety technological development tools to maximize the potential of our project.

What's next for FB.GenerateCloud

We would love to add the features that we couldn't implement due to the time limit. Our project has the potential of not only providing more services related to Facebook, such as interest-matching and finding event buddies, but also the potential of provide service cross more platforms including and not limited to instagram, twitter, and many more. Along with adding more services, we can also incorporate the use of wit.ai into the project, providing the users the option of making their requests in natural conversational sentences. By training the AI to understand the intention of the users, we can then request and g

We also plan to implement Pillow so that users can select what spahe they want their word cloud to appear in, as well as a Download as PDF function.

Share this project:

Updates