SnowPro Advanced Data Scientist

By Snowflake · Question Mode

QuestionQ37

VerifiedTrain and use machine learning models

A Data Scientist is building a two-stage machine-learning process for fraud detection. The process must first group customers with comparable buying behavior into peer groups, and then attempt to identify abnormal purchase behavior relative to a customer’s peer group.

What should the Data Scientist do?

Explanation

Customer peer groups are clusters of customers with similar behavior, which is an unsupervised-learning task. Derived normal and abnormal purchase examples provide labels for a subsequent supervised fraud classifier. AWS documents that clustering algorithms are unsupervised and that supervised learning trains on labeled examples.

Learn more

Community Discussion

No comments yet. Be the first to start the discussion!