Inspiration
On 19 June 2019, Chennai city, a city in India, officials declared that "Day Zero", or the day when almost no water is left, had been reached, as all the four main reservoirs supplying water to the city had run dry. People had to walk many, many kilometers to get water to drink. This later came to be known as ‘The 2019 Chennai Water Crisis’. Chennai isn’t alone. Major cities all around the world face water shortages during summer. Millions of people die due to water shortages each year. According to UNICEF [1], 2/3rd of the entire world faces severe water scarcity for at least 1 month per year. The credit of this water shortage usually goes to poor planning and management of the water that is received during monsoon each year. I have myself faced this shortage of water many times in my hometown. I am extremely interested in AI and for this hackathon, came up with a solution to this problem with an AI model.
What it does
The AI model predicts the water level of the reservoir Poondi, in Chennai, given the date
How we built it
I built this AI model using python and various libraries such as pandas, sns, etc
Challenges we ran into
The biggest challenge by far was to get the idea, and create this AI model. I was unable to get an inspiration for a good idea, and had to create this model in only a few hours, which was extremely hard and challenging for me, especially as I am in a different timezone and got confused with the deadline time. I also just couldn't attach it to an web app, due to the time constraints
Accomplishments that we're proud of
I am extremely proud of creating this AI model in such less amount of time. The model also works with 95% accuracy, which is amazing!
What we learned
I had to use decision trees to make this AI model which taught me some new topics, and I also had to work around some other bugs and issues that crept up.
What's next for Project Jal
The accuracy of the model can be increased. It can also be easily attached to a webapp and be used with better UI. The model can also be trained on the data of other reservoirs, which can enable it to be used world wide
Refrences
[1] - link
Log in or sign up for Devpost to join the conversation.