Explanation:
The correct answer is B. Using BigQuery ML to train a matrix factorization model requires the least amount of coding compared to other options. BigQuery ML provides high-level APIs for machine learning tasks directly within the BigQuery environment, reducing the need for extensive coding and data movement. Matrix factorization is a commonly used collaborative filtering technique for recommendation systems, making it suitable for recommending new games to users based on gameplay time data, user metadata, and game metadata.
Ultimate access to all questions.
You work at a gaming startup that has accumulated several terabytes of structured data stored in Cloud Storage. This data encompasses gameplay time data, user metadata, and game metadata. Your objective is to create a recommendation model to suggest new games to users. Given your primary goal is to minimize the amount of coding required, what approach should you take?
A
Load the data in BigQuery. Use BigQuery ML to train an Autoencoder model.
B
Load the data in BigQuery. Use BigQuery ML to train a matrix factorization model.
C
Read data to a Vertex AI Workbench notebook. Use TensorFlow to train a two-tower model.
D
Read data to a Vertex AI Workbench notebook. Use TensorFlow to train a matrix factorization model.
No comments yet.