01
True or false
1 point
A model uses a home's floor area and bedroom count to estimate its sale price. This is a regression task.
Free Practice Quiz Question List
1 Introduction to Machine Learning Online Quiz Questions
Use this free practice quiz with 20 questions to review 1 Introduction to Machine Learning, test your knowledge, and prepare for your next test or exam.
20 questions
02
True or false
1 point
A retailer groups shoppers using purchase patterns but has no predefined group labels. This is a supervised-learning task.
- A
True
- B
False
03
Choose one
1 point
A home-price dataset includes floor area, bedroom count, and the known sale price for each home. During supervised training, what role does the sale price play?
- A
A feature, because it is recorded for each home
- B
A label, because it is the value being predicted
- C
A cluster, because it identifies a group of homes
- D
A metric, because it measures the model's errors
04
Fill in the blank
1 point
A training table contains each home's floor area and its known sale price. The sale price is the .
05
Written response
1 point
A screening system must especially avoid missing people who have a condition. Which classification metric is particularly relevant?
06
Choose one
1 point
A team trains a model on emails labeled “spam” or “not spam” and wants it to assign one of those categories to new emails. Which problem type best fits?
- A
Regression, because the system assigns one of two possible outputs
- B
Classification, because the system assigns an email to a category
- C
Clustering, because the system predicts a known category
- D
Regression, because the system uses examples with labels
07
True or false
1 point
A team repeatedly checks its test-set score and changes the model in response. This can make the test set a less trustworthy final evaluation.
- A
True
- B
False
08
Fill in the blank
1 point
A model discovers groups of shoppers from purchase patterns without supplied group labels. This is a form of .
09
Choose one
1 point
During training, a model's performance on training data keeps improving, but its validation performance worsens. What is the most likely concern?
- A
The model may be overfitting the training examples
- B
The model is necessarily generalizing better
- C
The validation set has become the training set
- D
The task must be clustering rather than supervised learning
10
Choose all
1 point
A team has separate training, validation, and test sets. Which actions are appropriate? Select all correct choices.
- A
Use validation results to compare models or tune choices during development
- B
Repeatedly use test results to decide how to tune the model
- C
Reserve the test set for a final evaluation of the selected model
- D
Train the model on both sets before measuring its final performance
11
Written response
1 point
A retailer has unlabeled customer purchase histories and wants a model to discover groups of customers with similar patterns. What machine-learning problem type fits this goal?
12
Open ended
1 point
A home-price team wants large prediction errors to count more heavily and wants an evaluation that better indicates performance on new homes. Which regression metric should it consider, and what kind of examples should it use for evaluation?
13
Choose all
1 point
Before training a home-price model, a team audits its dataset. Which findings are data-quality problems that may undermine the model's usefulness? Select all correct choices.
- A
Some examples have missing feature values
- B
The dataset contains a mix of numerical and categorical features
- C
Some examples have been duplicated
- D
Each row represents one example
14
Choose one
1 point
A home-price model takes bedroom count and floor area as inputs and predicts sale price. How should bedroom count be classified in this setup?
- A
Bedrooms are labels because they are numerical counts
- B
Bedrooms are features because the model uses them as inputs
- C
Bedrooms are targets because the model predicts sale price
- D
Bedrooms are validation data because they are recorded for each home
15
Choose one
1 point
A retailer wants to group shoppers with similar purchasing patterns, but has no predefined shopper categories or labels. Which machine-learning problem best fits this task?
- A
Regression
- B
Clustering
- C
Classification
- D
Supervised learning
16
Written response
1 point
In a supervised dataset of homes, the model is trained to predict each home's sale price. What is the sale price called in this dataset?
17
Choose one
1 point
A team transforms its examples before training a model. Which practice best follows the recommended data-preparation workflow?
- A
Transform training examples but leave later examples untransformed.
- B
Transform only examples that the model predicts incorrectly.
- C
Apply needed transformations consistently to the relevant data.
- D
Avoid transforming any data, even when preparation is needed.
18
Choose one
1 point
An email model estimates the probability that each message is spam, then assigns either “spam” or “not spam” using a decision threshold. What kind of problem is it solving?
- A
Classification
- B
Regression
- C
Clustering
- D
Dimensionality reduction
19
Written response
1 point
Which regression metric measures the average size of prediction errors?
20
Choose one
1 point
A medical screening classifier should avoid missing people who have a condition. Which classification metric is especially relevant to this priority?
- A
Accuracy, because it is always the most useful classification metric.
- B
Mean squared error, because it penalizes larger numerical prediction errors.
- C
Precision, because it measures the fraction of predicted-positive cases that are actually positive.
- D
Recall, because missing positive cases is an important concern.