ironhack-labs / lab-customer-analysis-round-1

Lab | Customer Analysis Round 1

Remember the process:

1. Case Study
2. Get data
3. Cleaning/Wrangling/EDA
4. Processing Data
5. Modeling
6. Validation
7. Reporting

Abstract

The objective of this data is to understand customer demographics and buying behavior. Later during the week, we will use predictive analytics to analyze the most profitable customers and how they interact. After that, we will take targeted actions to increase profitable customer response, retention, and growth.

For this lab, we will gather the data from 3 csv files that are provided in the files_for_lab folder. Use that data and complete the data cleaning tasks as mentioned later in the instructions.

Instructions

• Read the three files into python as dataframes

• Show the DataFrame's shape.

• Standardize header names.

• Rearrange the columns in the dataframe as needed

• Concatenate the three dataframes

• Which columns are numerical?

• Which columns are categorical?

• Understand the meaning of all columns

• Perform the data cleaning operations mentioned so far in class

  • Delete the column education and the number of open complaints from the dataframe.
  • Correct the values in the column customer lifetime value. They are given as a percent, so multiply them by 100 and change dtype to numerical type.
  • Check for duplicate rows in the data and remove if any.
  • Filter out the data for customers who have an income of 0 or less.

lab-customer-analysis-round-1's Issues

missing fields in the csv files create problems in the next rounds

There are missing fields in the csv files ( e.g. Response , Sales Channel) that are needed for the next rounds in the customer analysis case study . What happens is that the students work on the data cleaning tasks using these csv files and afterwards they have to redo all their work because they receive new files with new fields in the next rounds, which creates a frustration to them because they have to redo and recheck all their cleaning code on the new file.

The solution: is to make sure that the same files that students work on from the first round can be used(including all fields) in later rounds so they can apply all the data cleaning and exploratory data analysis in one consistent pipeline.