In my previous blog post I discussed how to predict probabilities using a single label, and the link to this post can be found below:- https://tracyrenee61.medium.com/predicting-probabilities-with-python-b456334b85c6

In this post, I intend to cover the complexities involved in prediction probabilities of a multiple label dataset and walk through a real life example to show how this can be accomplished. I used the DrivenData competition question, Flu Shot Learning: Predict H1N1 and Seasonal Flu Vaccines, and the datasets needed have been saved in a repository of my github account, found below:- https://github.com/TracyRenee61/Swine-Flu

The problem statement for this competition question states:-

“Your goal is to predict how likely individuals are to receive their H1N1 and seasonal flu vaccines. Specifically, you’ll be predicting two probabilities: one for h1n1_vaccine and one for seasonal_vaccine.

Each row in the dataset represents one person who responded to the National 2009 H1N1 Flu Survey.”

The first thing I did was to load the libraries onto the Google Colab file I had created for this exercise:-

#artificial-intelligence #machine-learning #python #data-science #predictive-analytics