We did the research and got to know that garbage is not recycled in the way it should be. In most of the developing countries, all the garbage collected from the houses is directly dumped in landfills that cause air pollution and land water contamination. And, facts about the garbage in the ocean was much more shocking, each year amount of waste that is thrown in the ocean is 2 times the size of the United States.

What it does

Useful Garbage is an application and a Whatsapp Bot that classifies the garbage based on its decomposition method and recyclability using Machine Learning. It motivates the user by showing the data in a graphical method so that the user can understand his role in making this environment a better place for humans as well as other plants and animals.

How we built it

We built it using Python, TensorFlow and Twilio. We used TensorFlow and Keras for the machine learning model and streamlit for making the web interface of the app. For making the bot we used Twilio TwinML and the same Machine Learning Model we used for the web app. We also used Ngrok for forwarding the ports from the server. Whenever the user makes a request on Whatsapp it will be sent to the server for recognition of the Garbage and the server will provide the output on WhatsApp.

Challenges we ran into

We were using Twilio for the first and are new to Machine Learning. We are from different timezones so that was an issue but we completed the work on time.

Accomplishments that we're proud of

We are happy that we have completed this project before the deadline. We learnt lots of new things together. We are proud of contributing a possible solution to help protect the environment.

What we learned

We learned how to hack, How to make Machine Learning Models and Use Twilio along with it, How to push ourselves on little sleep, How to learn, How to look things up, communicate and collaborate, We gained confidence and friends from participating in this hackathon

What's next for Useful Garbage

Make it available on more platforms.

Built With

Share this project: