Machine Learning vs Deep Learning — Learn free

Self-check

Questions 1–10 of 10

1. What kind of raw dataset do Machine Learning (ML) models normally work on?
Question 1A. Unstructured data onlyB. Structured or semi-structured dataC. Images and video onlyD. Audio recordings onlyE. Free-text documents only
2. What kind of raw dataset do Deep Learning (DL) models work on?
Question 2A. Structured data in rows and columnsB. Semi-structured data onlyC. Unstructured data, where no one knows what will flow inD. Pre-cleansed feature tables onlyE. Relational database tables only
3. In Machine Learning, what is a feature table?
Question 3A. A list of product features for marketingB. A configuration file that lists model hyperparametersC. A dashboard summarising model accuracyD. A table no different from a traditional table, but containing cleansed dataE. A catalogue of image files for training
4. Which example is used to contrast Machine Learning with Deep Learning?
Question 4A. Counting cars, bikes and buses, first from rows and columns and then from picturesB. Predicting stock prices from historical tradesC. Translating text between languagesD. Detecting fraud in credit card transactionsE. Recommending films to viewers
5. If pictures are given to a Machine Learning model to count cars, bikes and buses, what is the expected outcome?
Question 5A. It will count them accuratelyB. It will count only the carsC. It will convert the pictures to a feature table automaticallyD. It will produce a heat map insteadE. It will not be possible, because ML models normally work on structured or semi-structured data
6. Which type of model can recognise objects in pictures and determine whether each is a car, a bike or a bus?
Question 6A. Deep Learning modelsB. Traditional Machine Learning modelsC. Data Mining modelsD. Dimensional data modelsE. Relational data models
7. What do Deep Learning models do with the unstructured dataset before classifying it?
Question 7A. Discard it and request structured data insteadB. Dig out the required data to prepare a cleansed dataset in the shape of feature tablesC. Convert it into a data warehouseD. Send it to a human for manual labellingE. Store it unchanged in a data lake
8. Which measurements make it relatively easy for an ML model to count cars, bikes and buses in a structured dataset?
Question 8A. Colour, brand and priceB. Speed, fuel type and ageC. Width, length and heightD. Owner, registration and locationE. Pixel count, resolution and format
9. Which of the following is NOT true of Machine Learning models as described in the comparison with Deep Learning?
Question 9A. They use structured or semi-structured raw dataB. They rely on feature tables of cleansed dataC. They classify data to produce expected resultsD. They normally recognise objects directly from picturesE. They handle data in rows and columns format
10. After Machine Learning versus Deep Learning, what is named as the next stage of learning to understand?
Question 10A. Reinforcement learning and transfer learningB. Data warehousing and data miningC. Data classification and data clusteringD. Descriptive and prescriptive analyticsE. Supervised and unsupervised learning