Air Health

Prediction of Particulate matter 10 value of air pollution to help people with breathing problems. We used and compiled previous five years of data from open datagov.in for delhi. We compiled the data and converted it into a time series, cleaned it and then used LSTM model to predict the future values for air pollution.

Tools/Libraries used for Data analysis and machine learning -

• pandas
• numpy
• keras
• sklearn
• matplotlib

Tools/Libraries used for messenger chatbot -

• Django

Tools/Libraries used for landing Page -

• HTML
• CSS
• Javascript

Screeshots -

Analysis of nizamuddin, Delhi, IN

Analysis of pitampura, Delhi, IN

Screenshot of messenger bot which is giving the current pm10 value. We can also get the predictions as well.

Performance Townhall

The landing page is accessible at datazen.tech.

Notes

• facebook chatbot is not live now. you can link it to your own facebook chatbot to try it.
• the project was initially made for HackData event and the whole project was uploaded here on october 14,2017