CLOUD-DIGITAL-LEADER practice question 2 of 35

A vacation home rental organization wants to predict the popularity of properties in their upcoming busy season. They do not have a data science team, and want to use their in-house database administration skills to create a machine learning model. What should the organization do?

Full set: 411 questions
Answer options
A: Use custom training in Vertex Al with TensorFlow.
B: Use BigQuery ML and create models using SQL. Correct answer
C: Build a model in AutoML using labeled data.
D: Integrate pre-trained APIs into their application.
Correct Answer:
B. Use BigQuery ML and create models using SQL.
Explanation

BigQuery ML lets users create, train, evaluate, and predict with machine-learning models directly in BigQuery by writing familiar SQL. This fits a team that lacks dedicated data scientists but already has database–administration/SQL expertise. They can stay inside the data warehouse, avoid exporting data, and build a regression or classification model to forecast property popularity for the next season.

Why Incorrect

A. Custom training in Vertex AI requires writing TensorFlow/PyTorch code and managing training infrastructure—skills a non–data-science team typically lacks.
C. AutoML simplifies ML but still needs dataset preparation, labeling, and additional UI/SDK steps; it does not specifically leverage SQL skills already present.
D. Pre-trained APIs (Vision, Natural Language, etc.) provide fixed models; they cannot be trained to forecast property popularity from organization-specific tabular data.

References

1. Google Cloud, “BigQuery ML Overview,” Docs > BigQuery ML, Section: What is BigQuery ML? (“enables data analysts to build ML models in BigQuery using SQL”). https://cloud.google.com/bigquery-ml/docs/introduction

2. Google Cloud, “BigQuery ML Tutorial: Creating a regression model,” Docs, Steps 1–2 (training and prediction entirely with SQL).

3. Google Cloud, “Vertex AI: Custom training,” Docs, Overview paragraph 1 (“use when you need full control over training code and environment”).

4. Google Cloud, “Vertex AI: AutoML Tabular,” Docs, Before you begin (“requires labeled dataset; AutoML handles training and tuning”).

5. Google Cloud, “Pre-trained ML APIs,” Product overview (“ready-to-use models; no training with your data”).

Scroll to Top