Got Toxic?

Inspiration

We were very interested by Co:here’s natural language processing API. We thought it could be utilized to help people who are experiencing unhealthy relationships, as we’ve personally known friends and classmates who have gone through toxic relationships.

What it does

Our website takes in an input of a phrase or sentence that the user wants to be analyzed and, using Co:here’s NLP model, determines the “toxicity” of the input.

How we built it

We started with building a backend out of Python using Flask. We utilized the Co:here API to analyze inputs. Afterwards, we built the front end website that took a form input and submitted it to the Python backend. Then we modified the back end program to take the front end input and return a webpage that displays Co:here’s NLP model’s toxicity results.

Challenges we ran into

Many of the libraries for this program we never used before, such as Flask and Co:here. A big challenge was how to get the data submitted from the front end to the back end as we ran into many errors and had zero experience with web app development.

Accomplishments that we're proud of

We are very proud to present a finished product that works and with errors greatly diminished.

What we learned

We learned a lot about Flask, front end development, HTML forms, Co:here’s API, and web apps.

What's next for Got Toxic?

We plan to add helpful resources for people in need, and those who are potentially in threatening circumstances. We also want to make the webpage more dynamic and be able to present more data in the future.