Chicken-Tastic Forecast

The program at work

Inspiration

We were inspired by the issue of water consumption, as it’s one of the leading problems that the world faces today. One of the main causes is from livestocks. So we decided to target this issue by providing a possible solution to a fundamental pillar or water overconsumption: agriculture.

What it does

Our program predicts the weight of chickens after a given number of days based on their birth weight and the diet they were fed. This helps plan for the growth of chickens and the potential water consumption of chickens. The program is modular and can be adjusted by farmers to gain insights on the best diet for their chickens.

How I built it

We started by filtering the data to keep the initial weight and and final weight of the chickens. Then, we removed outliars (chickens who died after a few days). Finally, we applied a linear regression model to the data to predict the final weight based on the input (initial weight, time of growth, diet).

Challenges I ran into

We had no prior knowledge in machine learning. Everything was learned on the fly and from scratch.

Accomplishments that I'm proud of

This was the first AI project of every team member. We are extremely happy to have come up with something working and scalable.

What I learned

We learned that we could use more prebuilt libraries from sklearn to our advantage. At first, we tried to do most of the data separation into training and test set manually, but this was inefficient.

What's next for Chicken-Tastic Forecast

We want to apply it to the real world with much more data, so each farmer can use it to tailor the diet of their chickens. It would also be pertinent to implement a public system which shares the data with all farmers to encourage collaboration and data sharing for a strong and eco-friendlty agricultural industry. Moreover, the project can be used with any kind of farm to plan for the growth and consumption of other kind of living animals. It is versatile and modular.