MLS-C01 Question 126
Single answerA retail company wants to improve its customer retention by identifying customers who are likely to stop purchasing within the next three months. The company has historical customer data, including purchase history, demographics, and interaction patterns. As a machine learning specialist, how would you frame this business problem as a machine learning problem?
- A
Frame it as a regression problem to predict the exact number of purchases a customer will make in the next three months.
- B
Frame it as a classification problem to predict whether a customer is likely to stop purchasing (churn) or not within the next three months.
- C
Frame it as a clustering problem to group customers based on purchase patterns and identify which groups are more likely to churn.
- D
Frame it as a recommendation problem to suggest products to customers based on their historical purchasing behavior.
Show answer and explanation
Correct answer: B
Explanation
The business problem focuses on identifying customers who are likely to stop purchasing (churn) within a specific time frame. This requires predicting a binary outcome (churn or not churn), making it a classification problem. Framing the problem correctly ensures that the appropriate machine learning approach is used to address the business need.
- A. Incorrect.
This is incorrect because predicting the exact number of purchases doesn't directly address the problem of identifying customers who are likely to churn. The focus here is on predicting customer churn, not purchase volume.
- B. Correct.
This is correct because the goal is to determine whether a customer will churn (binary outcome: churn or not churn) within a specific time frame, which is a classic classification problem.
- C. Incorrect.
This is incorrect because clustering is an unsupervised learning approach, and the goal here is not to group customers but to predict a specific outcome (churn or not churn) for each customer.
- D. Incorrect.
This is incorrect because recommendation systems aim to suggest products or services to users, which is unrelated to predicting customer churn.