Microsoft Azure AI Fundamentals (AI-900)Describe fundamental principles of machine learning on AzureMedium
A retail company wants to analyze customer purchasing behavior to identify distinct groups of customers for targeted marketing campaigns. They have a large dataset of past transactions but no predefined labels for customer segments. Which machine learning approach is most suitable for this task?
- AUnsupervised learning for clustering
- BSupervised learning for regression
- CReinforcement learning
- DSupervised learning for classification
Show answer & explanationAnswer & explanation
Correct answer: A. Unsupervised learning for clustering
Since the company has no predefined labels for customer segments, they need an unsupervised learning approach to discover inherent patterns and group similar customers together. Clustering is the specific unsupervised technique designed for this purpose.
Why the other options are wrong
- B. Supervised regression requires labeled data to predict continuous values, which is not the goal.
- C. Reinforcement learning is for agents learning actions in an environment, not for discovering data patterns.
- D. Supervised classification requires labeled data to predict predefined categories, which is not available here.
Unsupervised Learning (Clustering)
A type of machine learning that finds hidden patterns or data groupings in unlabeled datasets, with clustering being a primary technique.
- No labeled output variable is provided during training.
- Aims to discover structures and relationships within data.
- Clustering algorithms group similar data points together.
Memory trick: Unsupervised finds 'unseen' groups in purchase data.