Legalese is something that everyone's had to deal with, but it's often quite difficult and inaccessible. We find that even as students who have the privilege of getting an education, we cannot usually comprehend legal documents in their entirety because of all the jargon thrown in. So, we wanted to create a resource for everyone, especially since getting help with legal matters is usually expensive and not everyone has the money to afford it. We're hoping that this resource can help level the playing field in legal negotiations and give people of all backgrounds the ability to advocate for themselves in legal processes.
What it does
Our project is a legalese translator website. We provide a way for someone to translate their document through text input. We also allow individuals to create an account on our website to save all their translations in case they may need them again in the future.
How we built it
Since LegalEase is a website, we used HTML and CSS for the front-end. We used flask and python for the back-end, neither of which any of us had any experience with, and we used the Google Cloud translation API to train our ML model to convert legal jargon into common language.
Challenges we ran into
We were learning a lot of new technologies at the same time, and it was difficult to figure out python and flask at the same time. We also accidentally trained our ML model to translate into French at first, and had to quickly go to plan B after waiting multiple hours for it to train the model from our data.
Accomplishments that we're proud of
None of us have ever created an ML model before, so we're very proud of our end model from the Google Cloud API. We also truly believe we have created a project that can help make legal processes more accessible to people of all backgrounds, and even know this is a premature version, we're proud that this could potentially have a major impact.
What we learned
Through this project, we learned how to use flask and python, and we learned how to create an ML model.
What's next for LegalEase
We want to build our set of training data, and work on our ML model to make it even more accurate in its translations. After more testing, we want to roll this out and see if it can help people in lower-income communities.