1. You are designing a real-time anomaly detection system for a streaming data source in Amazon Kinesis Data Analytics. You want to use machine learning to detect anomalies and trigger alerts when they occur. Which of the following machine learning algorithms is best suited for this use case?

A) Principal component analysis (PCA)
B) Support Vector Machines (SVM)
C) Logistic regression
D) Random Cut Forest (RCF)



2. Which AWS service would be most appropriate to perform Extract, Transform, Load (ETL) jobs for a large amount of data stored in Amazon S3, with the requirement to handle streaming data and near-real-time data processing?

A) AWS Glue
B) Amazon Kinesis Data Firehose
C) Amazon Redshift
D) AWS Data Pipeline
E) Amazon EMR


3. Which of the following techniques is commonly used in natural language processing (NLP) to transform words into a continuous vector space in order to capture semantic meaning?

A) Word2Vec
B) Principle Component Analysis (PCA)
C) None of the above
D) Latent Dirichlet Allocation (LDA)
E) One-hot encoding


4. Which of the following statements is true about the accuracy metrics used in Amazon Forecast?

A) Symmetric Mean Absolute Percentage Error (SMAPE) is used to measure the accuracy of forecasted time series data
B) Mean Absolute Percentage Error (MAPE) is the only accuracy metric available in Amazon Forecast
C) Root Mean Squared Error (RMSE) is a percentile-based accuracy metric that is used to measure the error in the forecast of time series data
D) The weighted quantile loss (WQL) is a common accuracy metric that is used to measure the error in the forecast of time series data



5. When launching an Amazon EKS cluster, which of the following is a best practice for ensuring high availability?

A) Use a managed node group with a single instance in a single Availability Zone
B) Launch the EKS cluster in a single Availability Zone
C) Use a single instance type for all worker nodes
D) Deploy worker nodes across multiple Availability Zones
E) Use a self-managed node group with a single instance in a single Availability Zone