What it does

Our app is an API endpoint which allows other developers to hook into our harassment algorithm. We also built a chrome extension to show the algorithms performance and intelligence.

How we built it

We built an API in python flask which connected to a Postgresql database. We decided to use an object relational mapper to integrate with our database. We used python's natural language toolkit in conjunction with Google's bad words to train our model.

Challenges we ran into

The biggest challenge was our skill gap, we were passionate about learning something new and wanted to come into the project with an open mind. All of the team learned something new.

Accomplishments that we're proud of

Using NLP was a huge challenge for our project, but we are really proud of our algorithm and chrome app. We learned a lot of new skills that we wouldn't have through of learning outside the algorithm.

What we learned

We didn't know anything about natural language processing before we started this project. We found a research paper that detailed methods with which to recognise online harassment.

One of our teammates had never used python before so he was able to learn in trial by fire.

What's next for HarassBlock

We can index more of the web and gather more data in order to provide services that may be useful to advertising companies who wish to avoid being affiliated with negative websites. We could also see our API and algorithm being used on online forums and comment sections to minimize potential harassment and hate speech.

Share this project:


posted an update

HarassBlock should be ready to demo soon. We are very proud of what we have achieved and cannot wait to officially release it to the Chrome Extensions Store. The quicker our app is available to the public the sooner we can help society and the more users it had the more accurate and beneficial it will become. Thanks!

Log in or sign up for Devpost to join the conversation.