1. What AWS machine learning (ML) application service would be the best fit for a company looking to train and deploy an anomaly detection model for monitoring a large fleet of IoT devices in real-time, given the requirement of low-latency response times and cost-effectiveness?

A) Amazon SageMaker
B) AWS DeepLens
C) Amazon Kinesis Data Analytics
D) Amazon Fraud Detector
E) Amazon QuickSight


2. A financial services company wants to build an ML model that can detect fraudulent transactions in real-time using a combination of structured and unstructured data. Which AWS ML application service would be most appropriate for this use case?

A) Amazon SageMaker
B) Amazon Comprehend
C) Amazon Rekognition
D) Amazon Fraud Detector
E) Amazon Translate


3. Which of the following pre-processing techniques is best suited for handling noisy and misspelled text data in natural language processing (NLP) tasks?

A) Stopword removal
B) Stemming
C) Lemmatization
D) Spell checking
E) None of the above


4. Which of the following techniques is commonly used in natural language processing (NLP) to group similar words together based on their meaning, and can be used to identify synonyms and antonyms?

A) Latent Dirichlet Allocation (LDA)
B) Naive Bayes Classifier
C) K-Means Clustering
D) WordNet
E) None of the above


5. Which of the following statements is true regarding the use of Notebooks and IDEs in AWS SageMaker?

A) SageMaker Studio provides a built-in debugger that supports debugging of TensorFlow and PyTorch models.
B) Jupyter Notebooks in SageMaker do not support the integration of third-party libraries.
C) SageMaker Notebook instances provide an integrated dashboard for monitoring model training and evaluation.
D) SageMaker Studio provides a built-in visualization tool for data exploration and analysis.