Do we want to keep the gender feature?

UBC-MDS / Telco_Customer_Churn_Prediction_Group12

Data Analytics project performed by Group 12 for DSCI 522

MIT License

2 stars 7 forks source link

Do we want to keep the gender feature? #24

Closed jcasoli closed 2 years ago

jcasoli commented 2 years ago

I think we should all have some input here...

My vote would be to drop it, as I think using this feature could lead to some unintended consequences down the line. Please let me know what you think!

Anupriya-Sri commented 2 years ago

Agreed. The churn data and features impacting churn are used by companies to develop the marketing strategy and released customized offer. By including gender as a feature, we would add a bias to the models which may not be correct. So, we should drop this column for transformation and predictions.

jcasoli commented 2 years ago

the group has agreed to drop the gender feature in the preprocessing script.

zzhzoe commented 2 years ago

Agree with drop the gender feature in the preprocessing script.