1. A company wants to build a machine learning model to predict customer churn for their e-commerce platform. The company has historical data on customer behavior and purchases stored in a relational database. Which AWS service can be used to ingest and preprocess the data for machine learning training?

A) Amazon Redshift
B) Amazon SageMaker
C) Amazon Aurora
D) Amazon Kinesis Data Streams
E) AWS Data Pipeline


2. Amazon EFS provides a scalable and fully managed file storage service that can be accessed by multiple compute instances concurrently.

A) Amazon EFS provides a scalable and fully managed file storage service that can be accessed by multiple compute instances concurrently.
B) Amazon EFS is optimized for high-performance, low-latency storage access, making it the ideal choice for latency-sensitive machine learning workloads.
C) Amazon EFS provides built-in data encryption at rest and in transit, ensuring data privacy and compliance.
D) Amazon EFS is compatible only with Amazon EC2 instances and cannot be used with other compute services.
E) Amazon EFS is the most cost-effective storage option for machine learning workloads that require high-performance file storage.


3. Which of the following statements is true regarding the use of Amazon Forecast in forecasting demand for a product?

A) Amazon Forecast requires a high degree of expertise in machine learning, making it difficult for non-technical users to utilize.
B) Amazon Forecast uses a fixed, pre-defined algorithm that cannot be customized to suit the specific needs of a business.
C) Amazon Forecast can incorporate a wide range of data sources, including historical sales data, external factors, and seasonality, to provide accurate demand forecasting.
D) Amazon Forecast can only be used for single-item forecasting and cannot be scaled to handle large product catalogs.
E) Amazon Forecast is only suitable for short-term forecasting and cannot be used for long-term forecasting.


4. You want to set up automatic scaling for an AWS resource based on a CloudWatch metric. Which of the following statements about CloudWatch alarms and scaling policies is true?

A) CloudWatch alarms can only trigger scaling policies for resources in the same AWS region.
B) CloudWatch alarms can only trigger scaling policies based on average or maximum metric values.
C) CloudWatch alarms can trigger scaling policies that increase or decrease the number of instances or capacity of AWS resources.
D) CloudWatch alarms and scaling policies can only be used with Amazon EC2 instances.



5. Which feature of Amazon Polly enables developers to customize and control the pronunciation of specific words or phrases?

A) NTTS voices
B) Pronunciation Lexicons
C) Speech Synthesis Markup Language (SSML)
D) Neural TTS