Timedora
Free Practice Quiz Question List

3 Classification Fundamentals Online Quiz Questions

Use this free practice quiz with 20 questions to review 3 Classification Fundamentals, test your knowledge, and prepare for your next test or exam.

20 questions
01
True or false
1 point

A classifier is trained using examples whose correct classes are already known. Is this supervised learning?

  1. A

    True

  2. B

    False

02
Choose one
1 point

A model must assign each input to one of four plant species. Which type of classification is this?

  1. A

    Binary classification

  2. B

    Multiclass classification

  3. C

    Unsupervised classification

  4. D

    Feature classification

03
Fill in the blank
1 point

In a labeled training example, the known answer paired with the input is the .

04
Written response
1 point

What are the categories that a classification model assigns to inputs called?

05
True or false
1 point

When making predictions for an unseen email, the model should be given the email's known spam/not-spam label as an input feature.

  1. A

    True

  2. B

    False

06
Choose one
1 point

An email classifier predicts whether a message is spam. Which item is a valid input feature rather than the target label?

  1. A

    Whether the email is spam

  2. B

    The correct spam/not-spam answer supplied to the model

  3. C

    The number of words in the email's subject

  4. D

    The email's known label

07
Choose all
1 point

Select all practices that support a valid evaluation of a classifier on new examples.

  1. A

    Fit the model using labeled examples.

  2. B

    Keep some labeled examples separate from training for evaluation.

  3. C

    Use test labels to compare with predictions, but not as model input features.

  4. D

    Include each test example's correct label among the model's input features.

08
Fill in the blank
1 point

To estimate how well a classifier works on new examples, evaluate it on labeled data kept separate from training. This held-out data is the .

09
Choose one
1 point

A screening system must avoid missing positive cases, even if it sometimes raises false alarms. Which measure is most directly relevant to that priority?

  1. A

    Accuracy

  2. B

    Precision

  3. C

    Recall

  4. D

    The number of features

10
Written response
1 point

A classifier makes 24 positive predictions, of which 18 are correct. What is its precision as a percentage? Enter the percentage-point value.

11
Choose all
1 point

Select all statements that correctly interpret outcomes in a binary classifier's confusion matrix.

  1. A

    Predicted positive, actually positive: true positive.

  2. B

    Predicted negative, actually positive: false negative.

  3. C

    Predicted positive, actually negative: false positive.

  4. D

    Predicted negative, actually negative: false positive.

  5. E

    A false negative is a positive prediction that is incorrect.

12
Open ended
1 point

Explain why accuracy alone can give a misleading impression when one class is much more common than another. Distinguish how precision and recall help assess different kinds of errors.

13
Choose one
1 point

Which outcome is a false positive for an email spam classifier?

  1. A

    A spam email correctly flagged as spam

  2. B

    A non-spam email incorrectly flagged as spam

  3. C

    A spam email incorrectly left unflagged

  4. D

    A non-spam email correctly left unflagged

14
Choose one
1 point

An email model assigns each message to one of two categories: spam or not spam. What type of classification task is this?

  1. A

    Multiclass classification

  2. B

    Binary classification

  3. C

    Unsupervised classification

  4. D

    Regression

15
True or false
1 point

A classifier is more likely to work well when its training examples represent the task and its input features will be available at prediction time.

  1. A

    True

  2. B

    False

16
Choose one
1 point

An email classifier is being prepared to predict whether a message is spam. Which item is an input feature rather than the label?

  1. A

    Whether the email contains a link

  2. B

    The known category spam or not spam

  3. C

    The correct category used to score the prediction

  4. D

    The answer the model is being trained to predict

17
Choose one
1 point

A team wants to estimate how well its classifier will perform on new emails. What should it do with a separate set of labeled examples?

  1. A

    To add more examples to the training set

  2. B

    To provide the model with the labels it should predict

  3. C

    To estimate performance on new examples by comparing predictions with known labels

  4. D

    To ensure that every prediction is correct

18
Choose one
1 point

A message is actually spam, but a classifier predicts that it is not spam. In a binary confusion matrix, how is this result classified?

  1. A

    True positive

  2. B

    False positive

  3. C

    True negative

  4. D

    False negative

19
Written response
1 point

On a test set of 100 cases, a classifier records 18 true positives, 63 true negatives, 7 false positives, and 12 false negatives. What is its accuracy as a percentage?

20
Written response
1 point

A classifier identifies 21 of 30 actual positive cases in a test set. What is its recall as a percentage?