Mastering Marketing Data Science

EXERCISE 2.1: DATA CLEANING AND TRANSFORMATION

Objective: Clean and transform a dataset to prepare it for analysis.

Tasks:

1. Handle missing values (NaN) in the “data_cleaning_transfomation.csv” dataset.
2. Convert the ‘Last Login Date’ from a string to a datetime object.
3. Create a new feature, ‘Monthly Spend per Day’, by dividing ‘Monthly Spend’ by ‘Subscription Length’.

Steps:

1. Importing Required Libraries:

   1. import pandas as pd

   • pandas is used for data manipulation and analysis.
2. Loading the Data:

   2. data_exercise_1 = pd.read_csv('path_to_csv_file')

   This line of code reads the CSV file containing the data into a Pandas DataFrame, enabling us to work with the data in Python.

3. Handling Missing Values:
   • Filling Missing ‘Age’ Values:

   3. mean_age = data_exercise_1['Age'].mean()
   4. data_exercise_1['Age'].fillna(mean_age, inplace=True)

   Here, we calculate the mean of the ‘Age’ column and fill missing values (NaN) in the ‘Age’ column with this mean. This approach is chosen as age data typically follows a normal distribution, making the mean a good estimate for missing values.

   • Filling Missing ‘Monthly Spend’ Values:

   5. median_monthly_spend = data_exercise_1['Monthly Spend ($)'].median()
   6. data_exercise_1['Monthly Spend ($)'].fillna(median_monthly_spend, inplace=True)
   

   We fill missing values in ‘Monthly Spend ($)’ with the median, because financial data often has outliers, and the median is less sensitive to them compared to the mean.

   • Filling Missing ‘Feedback Score’ Values:

   7. mode_feedback ...