CyberSafe

Protecting you against online harassment and bullying

Comment

This project was created for High Tech Hacks 2022 by Team Wizards

Inspiration

Millions of individuals are harassed and bullied online, and the problem is sure to worsen as more people use the internet. Users are not well-protected on social media platforms. Furthermore, according to a Pew Research Center report, 59 percent of American students have been harassed or bullied online. As programmers, we must strive to create software that makes the internet safer for everyone.

What it does

It aims to prevent mental health disorders such as depression and anxiety caused by online harassment by predicting whether online messages indicate bullying or not.

How we built it

We used Bootstrap, HTML/CSS, Express, React, Node, and Javascript for the frontend, and we used Google Collab, Google Cloud, Tensorflow, BERT (Bidirectional Encoder Representations from Transformers), Flask for the backend

Challenges we ran into

  • Managing our scope such that we delivered a genuine product, even if it was a bit bare bones
  • CSS and website design
  • Working in different time zones: team members from west, east coast of America, and India
  • Utilizing the backend, we got it all to work in the end.
  • Finding Dataset: Due to limited Mental Health data as patient information is confidential, we were not able to find many datasets and it was only after a lot of browsing that we found the Bullying Traces Dataset
  • Twitter's Enhanced Developer Mode only allows access to recent tweets and our algorithm was only allowed to access 900 tweets in 15 minutes so it was difficult to train and the algorithm cannot be rerun multiple times due to this limit (We could not show multiple predictions in the demo due to this reason.)

Accomplishments that we're proud of

We're happy to have created a valuable product that helps combat online abuse and cyberbullying while appealing to a broad audience. We were proud of what we had accomplished in such a short time.

What we learned

This weekend taught us a lot. We discovered —

  • Use Flask to create fast, scalable, and simple web apps.
  • How utilizing Flask allows us to construct backends more quickly.
  • Using sentiment analysis (BERT)
  • How to preprocess data from CSV files
  • Learning how to use MERN (Mongo DB, Express, React, Node) Stack and passing data from frontend to backend
  • Using Twitter's Tweepy API

Finally, we learned a lot from one another, and High Tech hacks during the weekend.

What's next for CyberSafe

We aim to develop this platform and collaborate with users to make it a genuine free product that benefits everyone and introduces new features and enhancements.

Sources

Share this project:

Updates