AI-900 Question 84
Single answerA company wants to use machine learning to predict customer churn based on historical customer data. Which of the following techniques is most suitable for this scenario?
- A
Classification
- B
Clustering
- C
Regression
- D
Reinforcement Learning
Show answer and explanation
Correct answer: A
Explanation
The goal of the scenario is to predict whether a customer will churn or not, making it a classification problem. Classification is a supervised learning approach that is specifically used to predict discrete labels or categories, such as 'churn' or 'no churn,' based on historical labeled data.
- A. Correct.
Classification is the correct choice as predicting customer churn involves categorizing customers into two classes (e.g., 'will churn' or 'will not churn'). It is a supervised learning technique designed for such binary or multi-class prediction tasks.
- B. Incorrect.
Clustering is incorrect in this case. Clustering is an unsupervised learning technique used to group data into clusters based on similarity, rather than predicting specific outcomes such as churn.
- C. Incorrect.
Regression is used for predicting continuous numerical values, such as sales or temperature, not for categorical outcomes like churn.
- D. Incorrect.
Reinforcement Learning is a different paradigm of machine learning focused on decision-making in dynamic environments, such as training an agent to play a game. It is not suitable for this static prediction task.