For this lab, we still keep using the marketing_customer_analysis.csv
file that you can find in the files_for_lab
folder.
Remember the previous rounds. Follow the steps as shown in previous lectures and try to improve the accuracy of the model. Include both categorical columns in the exercise. Some approaches you can try in this exercise:
- use the concept of multicollinearity and remove insignificant variables
- use a different method of scaling the numerical variables
- use a different ratio of train test split
- use the transformation on numerical columns which align it more towards a normal distribution
We are using the marketing_customer_analysis.csv
file.
Already done in rounds 2 to 7.
Bonus: Build a function, from round 2 and round 7, to clean and process the data.
Done in the round 3.
Description:
- Try to improve the linear regression model.