Associate Data Practitioner · Free Practice Question Medium
Question 27
A data analyst wants to use a pre-trained language model for sentiment analysis on customer reviews directly within BigQuery. Which approach should they use?
-
A
Export the data to Vertex AI then import results back
-
B
Use BigQuery ML with remote model connection to Vertex AI
-
C
Create a custom sentiment analysis model in BigQuery ML
-
D
Run sentiment analysis in Python and load results to BigQuery
Reveal correct answer
Correct answer: B
Explanation
Option B is CORRECT. BigQuery ML's remote model connections allow you to use pre-trained models hosted in Vertex AI directly from BigQuery SQL queries. This feature enables sentiment analysis on customer reviews without moving the data out of BigQuery. You can create a connection to a pre-trained text sentiment model in Vertex AI and then call it using standard BigQuery ML functions, getting the sentiment scores as part of your query results.
Option A is INCORRECT. Exporting data to Vertex AI and importing results back creates a complex, multi-step process that's unnecessary given BigQuery ML's ability to connect to remote models. This approach would be less efficient and more difficult to incorporate into regular analytical workflows.
Option C is INCORRECT. Creating a custom sentiment analysis model in BigQuery ML would require extensive training data and ML expertise. It's unnecessary when pre-trained models with high accuracy are already available through remote connections.
Option D is INCORRECT. Running sentiment analysis in Python requires extracting data from BigQuery, processing it separately, and then loading results back. This creates a more complex pipeline compared to using remote models directly within BigQuery queries.
Discussion
Think the marked answer is wrong, or have a better explanation? Share it below — comments appear after review.
