1. Which AWS service can be used to collect and stream real-time data for machine learning models, providing sub-second latency with automatic scaling, and enabling data analysis and aggregation through built-in visualization tools?

A) Amazon DynamoDB
B) Amazon Kinesis Data Streams
C) Amazon Redshift
D) AWS Glue



2. A retail company wants to build a machine learning model to predict customer churn based on historical customer transaction data stored in an on-premises database. Which AWS service should the company use for ingesting this data for machine learning?

A) Amazon S3 Transfer Acceleration
B) AWS Glue
C) Amazon Kinesis Data Streams
D) Amazon AppFlow
E) AWS Snowball


3. Which of the following techniques is commonly used in natural language processing (NLP) to extract relevant information and relationships from text data, by identifying and classifying named entities such as people, organizations, and locations?

A) Named entity recognition (NER)
B) Convolutional neural networks (CNNs)
C) Latent semantic analysis (LSA)
D) None of the above
E) Decision trees


4. Which of the following is NOT a supported Amazon Rekognition image recognition feature?

A) Speech-to-text conversion
B) Object detection
C) Scene understanding
D) Text recognition
E) Face analysis


5. You need to monitor a set of resources in your AWS account and create alarms to alert you when specific conditions are met. Which of the following statements about Amazon CloudWatch alarms is true?

A) Amazon CloudWatch alarms can be created to notify you when a metric crosses a threshold you define.
B) Amazon CloudWatch alarms can only monitor metrics that are published to CloudWatch by AWS services.
C) Amazon CloudWatch alarms can only be created for resources in the same AWS region as the CloudWatch metrics.
D) Amazon CloudWatch alarms can only be created for EC2 instances and S3 buckets.