1. Which of the following approaches is commonly used in natural language processing (NLP) to represent words as high-dimensional vectors in order to capture their meaning and relationships?

A) None of the above
B) Support vector machines (SVMs)
C) Bag-of-words
D) Random forests
E) Long Short-Term Memory (LSTM) networks


2. You are working on a project that involves analyzing and visualizing customer data from multiple sources, including transaction records, social media interactions, and web logs. You want to create a centralized data repository that can store and process the data and provide real-time visualization capabilities. Which of the following AWS services would you use for this purpose?

A) Amazon Redshift
B) Amazon DynamoDB
C) Amazon Elasticsearch
D) AWS Glue
E) Amazon Elastic MapReduce (EMR)


3. You are training a deep learning model to classify images of animals. Which of the following activation functions is commonly used in the output layer of a multi-class classification model?

A) Sigmoid
B) Softmax
C) ReLU
D) LeakyReLU
E) Tanh


4. Which of the following is a correct statement about Amazon Comprehend's Custom Classification API?

A) Custom Classification API only supports text classification in English language.
B) Custom Classification API requires you to provide labeled training data in JSON format only.
C) Custom Classification API only works with predefined categories and it does not allow you to add new categories.
D) Custom Classification API enables you to train your own document classifier using your own training data.



5. What is the purpose of the 'entity type' in Amazon Fraud Detector?

A) It defines the type of entity that the detector will process, such as a customer or a transaction.
B) It defines the type of machine learning model that will be used to detect fraud.
C) It defines the output format of the fraud detection results.
D) It defines the format of the input data to be used in the detector models.