Building and Using Google Cloud AI and ML Solutions CLOUD-DIGITAL-LEADER Practice Questions (Page 3)

Questions 11–15

1. 11foundation · easy

   What is Cloud TPU?

   • AAn open-source machine learning framework.
   • BA managed service for deploying ML models.
   • CGoogle's proprietary hardware accelerator optimized for TensorFlow and ML workloads.
   • DA pre-trained API for image recognition.
   Reveal Solution Discussion Select an answer first
2. 12foundation · easy

   A data analyst wants to train a machine learning model to predict customer churn. The data is already stored in BigQuery. Which Google Cloud approach would allow the analyst to build and run the model without moving the data or writing Python code?

   • AUse the Natural Language API to analyze customer feedback and predict churn.
   • BExport the data to a CSV file and upload it to Vertex AI for AutoML training.
   • CDeploy a pre-trained Vision API model to classify customer images.
   • DUse BigQuery ML with SQL statements to create and evaluate the model.
   Reveal Solution Discussion Select an answer first
4. 13application · medium

   A global e-commerce platform wants to allow sellers to list products in their local language while automatically showing the listing to buyers in the buyer's preferred language. The platform already stores all product descriptions as text. They want a simple API-based solution that requires no ML model management. What should they use?

   • AUse the Speech-to-Text API to convert the descriptions to audio for buyers
   • BTrain a custom AutoML Translation model for each seller's language pair
   • CUse the Natural Language API to classify the product descriptions by language
   • DUse the Cloud Translation API to translate product descriptions into the buyer's language in real time
   Reveal Solution Discussion Select an answer first
5. 14application · medium

   A social media company needs to automatically flag user-uploaded images that contain inappropriate content before they are published. They also want to extract any text visible in the images (e.g., memes with text) for moderation review. They have no ML team and want a pre-built solution. What should they use?

   • AUse the Natural Language API to analyze the image captions
   • BTrain a custom AutoML Vision model to detect inappropriate content and read text
   • CUse the Vision API for safe-search detection and OCR on the images
   • DUse the Video Intelligence API to analyze the images as a video stream
   Reveal Solution Discussion Select an answer first
6. 15foundation · easy

   A financial services company wants to build a custom AI model to detect fraudulent transactions unique to its customer base. Why would they choose Gemini Enterprise Agent Platform (formerly Vertex AI) over a pre-trained API?

   • ATo use a model that is already trained on all possible fraud patterns.
   • BTo automatically comply with all financial regulations without additional configuration.
   • CTo avoid the need for any data storage or processing in the cloud.
   • DTo train a model on their proprietary transaction data, enabling more accurate fraud detection for their specific patterns.
   Reveal Solution Discussion Select an answer first