Classroot was inspired by the experience of one of our team members. In high school, Eleanor took a logic course in a classroom decorated with posters of famous mathematicians — well, famous white, male mathematicians. She learned what many girls in STEM know: if you can't see yourself reflected in your role models, it's hard to feel confident.

24 hours ago, we came together with the mission of addressing the gender gap in computer science. Made up of women who have experienced the issue and men who are enthusiastic allies, our team decided to help solve a rarely-addressed aspect of the problem: classroom design.

It’s not just our personal experience. Studies show that girls are more likely to pursue CS when they’re taught in neutrally decorated classrooms rather than stereotypically “nerdy” environments. Teachers who absolutely support girls in STEM might not realize that classroom environment matters a whole lot. Our goal is to provide a simple way to assess their classroom’s inclusivity — and just as importantly, start growing toward a more equitable pedagogy.

What it does

Classroot provides resources for teachers to create an all-gender inclusive environment for students by providing insights and suggestions on overall classroom environment, specific elements placed in the classroom, and curriculum.

It evaluates the inclusivity of the user’s classroom photo, uploaded to our web application, by looking at different elements identified through machine learning (such as colors, general props and objects placed in the room, or violence measure).

It also examines wall displays and assesses the diversity (both gender and ethnic backgrounds) of the figures included. (A photo of posters featuring only certain gender and/or ethnicity will prompt a gentle suggestion to improve.)

To encourage long-term, sustainable improvements, the Alexa app can be used on a daily basis to suggest class instruction ideas and send curricular reminders by text.

How we built it

Front end: We used Angular to develop a web application and have it connect to Firebase for data storage.

Back end: We used Clarifai API to custom-train some models and also utilize general models to predict and tag images through machine learning. First it was developed in Python; later, we wrote it again in JS to seamlessly incorporate it with the front end.

For Alexa, we used AWS lambda and Amazon Developer Console (node.js) in order to create an app. We also created a python-based Twilio messaging system so we can update the user with info.

Challenges we ran into

Lack of data set to train the models for tagging

Learning some documentations regarding Firebase, js, etc.

Accomplishments that we're proud of

We brought together a group of people with diverse skills and interests to make a really important project. Everyone, from our machine learning expert to our educational psych major, provided a valuable perspective.

What we learned

Learning how to quickly integrate frontend to backend, especially given the time limit.

What's next for Classroot

Our goal is to make inclusivity part of every teacher's daily routine. We will extend Alexa integration to provide to do lists and curricular insights - and make Classroot an even more helpful teacher's aide.

You can currently text ( 562-247-0818 ) for information provided by Alexa. In the future (and with users' consent), we will send the results of their classroom analyses right to their mobile devices.

By continuing to train our algorithm with our users' input, we can improve the accuracy of our training model. We're excited to see Classroot grow into a more accurate, useful, and inclusive product for everyone.

Share this project: