Timedora
Free Practice Quiz Question List

5 Classification Online Quiz Questions

Use this free practice quiz with 20 questions to review 5 Classification, test your knowledge, and prepare for your next test or exam.

20 questions
01
True or false
1 point

A model that predicts whether a message is spam or not spam is performing classification, while a model that predicts a house price is performing regression.

  1. A

    True

  2. B

    False

02
Choose one
1 point

A classifier assigns each support ticket to either “urgent” or “not urgent.” What kind of classification task is this?

  1. A

    Multiclass classification

  2. B

    Binary classification

  3. C

    Regression

03
Fill in the blank
1 point

A message classifier uses words in each message as to help predict its category.

04
Choose one
1 point

An email classifier labels a message as spam when its estimated spam probability exceeds 0.500.50. For a message with an estimated probability of 0.820.82, which label does it predict?

  1. A

    Spam

  2. B

    Not spam

  3. C

    A numerical quantity rather than a category

05
True or false
1 point

Every classifier must use a straight-line decision boundary, even when its feature space has many dimensions.

  1. A

    True

  2. B

    False

06
Written response
1 point

What is the term for the location in feature space where a classifier's predicted class changes?

07
Choose all
1 point

Select all statements that correctly describe how a model can form classification regions.

  1. A

    A decision tree divides feature space through feature-based splits.

  2. B

    A linear model must create an irregular boundary by consulting nearby labeled examples.

  3. C

    A k-nearest neighbors classifier can form an irregular boundary from nearby labeled examples.

  4. D

    A linear support vector machine always predicts by choosing the most common label among nearby examples.

08
Fill in the blank
1 point

An image model chooses among “cat,” “dog,” and “bird.” This is an example of .

09
Written response
1 point

Which classification model predicts the class most common among a new example's nearest labeled training examples?

10
Choose all
1 point

Select all models whose classification strategies are described as making feature-based splits or predicting from nearby labeled training examples.

  1. A

    Decision trees

  2. B

    k-nearest neighbors

  3. C

    Logistic regression

  4. D

    Support vector machines

11
Choose one
1 point

Which description best explains how logistic regression can produce a binary class prediction?

  1. A

    It predicts the most common class among nearby labeled examples.

  2. B

    It applies a sequence of feature-based rules to reach a class prediction.

  3. C

    It estimates class probabilities that can be converted into labels using a threshold.

  4. D

    It seeks a separating boundary, often aiming for a wide margin.

12
Open ended
1 point

Explain how a classifier uses training examples to predict a category for a new example, and describe what its decision boundary represents.

13
True or false
1 point

Some classifiers provide a score or probability for each class rather than outputting only a final label.

  1. A

    True

  2. B

    False

14
Choose one
1 point

An email classifier assigns the label “spam” when its estimated spam probability is above a chosen threshold. What can happen if the threshold is changed?

  1. A

    It changes the example's features.

  2. B

    It can change which examples receive the “spam” label.

  3. C

    It turns classification into a prediction of a numerical quantity.

  4. D

    It changes the model's training examples into unlabeled data.

15
Choose one
1 point

A model uses labeled examples of messages to predict whether each new message is spam or not spam. Which description best identifies this task?

  1. A

    It predicts a numerical quantity from the input features.

  2. B

    It assigns the example to a category based on its features.

  3. C

    It groups examples without using any labeled training examples.

  4. D

    It changes the features until they match a target value.

16
Choose one
1 point

On a plot of examples described by two features, what does the curve separating regions predicted as different classes represent?

  1. A

    The set of training examples that have the most common label.

  2. B

    A score that measures how many features an example has.

  3. C

    The location where the predicted class changes from one class to another.

  4. D

    The numerical quantity that a regression model predicts.

17
Written response
1 point

What is the classification term for the location in feature space where a model's predicted class changes?

18
Choose one
1 point

An email classifier reaches its label by applying a sequence of rules about the message's features. Which model type best matches this approach?

  1. A

    A decision tree

  2. B

    A k-nearest neighbors classifier

  3. C

    A logistic regression model

  4. D

    A support vector machine

19
Written response
1 point

What classification model predicts a new example's class from the labels of its nearest labeled training examples?

20
Choose one
1 point

A binary logistic regression model estimates a message's spam probability as 0.820.82 and uses a threshold of 0.500.50. Which label does it predict?

  1. A

    It predicts not spam because the probability is below 1.001.00.

  2. B

    It predicts spam because 0.820.82 exceeds the 0.500.50 threshold.

  3. C

    It cannot predict a label because logistic regression only produces features.

  4. D

    It predicts not spam because the threshold is lower than the probability.