4-1 Feature Selection for Predictive Modeling
Contains unread posts
Tiffany Rudman Quinn posted Nov 19, 2023 6:39 PM
Subscribed
Hi all!
After reviewing the requirements for Project Two, I am thinking I would use the
wrapper feature selection method as this method allows me to look at more than
one input features and then will help to determine the variable that will provide
the best model (Brownlee, 2020). In project one, I followed the code the
professor provided and determined variables that have a higher probability of
being a reason an employee is leaving the organization. The wrapper method
will also help to eliminate the variables that may be redundant and have no
effect on the employee attrition rate. We want to analyze the data to determine
what factors are causing employee to leave. This will allow the HR team to then
make changes within the organization to retain employees.
Once I have the variables selected, I can then use feature engineering to train a
predictive model and see what the best combinations of variables are that need
to be improved in order for employees to not leave the organization. I can then
use different combinations of the variables to get the best results. Being able to
change out the combination of the variables and train the model each time will
help to make HR understand what is happening and why.
Hope you all have a great holiday week!
Thanks,
Tiff
References:
Brownlee, J. (2020, August 20).
How to Choose a Feature Selection Method
For Machine Learning.
Machine Learning Mastery.
How to Choose a
Feature Selection Method For
Machine
Learning -
MachineLearningMastery.com.
less
