In this project, I focused on building a non-linear model to classify the conservation status based on: observations, park name, and category. The project can be found here .
My findings show that the number of observations impacts the conservation status the most. I’d like to hear some feedback on how I can improve the model.
Thanks,
Matteo
Woah, that’s special!
To be honest, I didn’t really know how/what you did at the second half of your code, but I think you built a machine learning algorithm that predicted the conservation status of each animal. And used this algorithm to predict the conservation status for the animals with the missing values.
I just finished the visualization part in the Data Science course, and I’ll be starting on the Natural Language Processing section. Hopefully I’ll get to where you’re at sometime soon 
Best wishes,
Safwan