wayground logo

Free Printable Worksheets

NEW

Font size

S
M
L
XL
Worksheets

Machine Learning & Deep Learning Quiz

Total questions: 50

Worksheet time: 25mins

Name
Class
Date
1.

Which of the following is a type of dataset used in Machine Learning?

a)

Training dataset

b)

Temporary dataset

c)

Background dataset

d)

Random dataset

2.

Continuous data refers to:

a)

Data that can only take fixed categories

b)

Data measured on an infinite scale

c)

Data with labels

d)

Data without labels

3.

Which of the following is labeled data?

a)

Collection of tweets without sentiment tags

b)

Images tagged as “cat” or “dog”

c)

Audio recordings with no labels

d)

Random survey responses

4.

The main purpose of a test dataset is to:

a)

Train the model

b)

Validate the hyperparameters

c)

Evaluate model performance

d)

Store unused data

5.

A flowchart for Machine Learning usually begins with:

a)

Model evaluation

b)

Data collection

c)

Feature engineering

d)

Hyperparameter tuning

6.

Supervised learning requires:

a)

Only unlabeled data

b)

Partially labeled data

c)

Fully labeled data

d)

Randomized data

7.

An example of unsupervised learning is:

a)

Linear regression

b)

K-means clustering

c)

Decision trees

d)

Logistic regression

8.

Which of the following is an example of reinforcement learning?

a)

Predicting house prices

b)

A robot learning to walk

c)

Identifying spam emails

d)

Customer segmentation

9.

Semi-supervised learning uses:

a)

Only labeled data

b)

Only unlabeled data

c)

Both labeled and unlabeled data

d)

No data

10.

Which ML model is prone to overfitting if not pruned?

a)

Linear regression

b)

Decision trees

c)

Naïve Bayes

d)

K-means

11.

Which algorithm works best for text classification?

a)

K-means

b)

Naïve Bayes

c)

KNN

d)

DBSCAN

12.

In SVM, the kernel trick is used for:

a)

Feature selection

b)

Handling non-linear data

c)

Overfitting

d)

Feature scaling

13.

Ridge regression uses:

a)

L1 regularization

b)

L2 regularization

c)

Dropout

d)

Bagging

14.

Which method combines multiple decision trees?

a)

Logistic regression

b)

Random forest

c)

Linear regression

d)

PCA

15.

Gradient boosting differs from random forests because:

a)

Uses bagging

b)

Builds trees sequentially

c)

Uses PCA

d)

Is unsupervised

16.

Feature scaling is important because:

a)

Prevents missing values

b)

Keeps all features on comparable scale

c)

Reduces dataset size

d)

Prevents bias in labels

17.

Which library is primarily for data visualization?

a)

NumPy

b)

Matplotlib

c)

TensorFlow

d)

Scikit-learn

18.

Which deep learning library is developed by Google?

a)

PyTorch

b)

TensorFlow

c)

Scikit-learn

d)

OpenCV

19.

Which library is widely used for computer vision tasks?

a)

TensorFlow

b)

OpenCV

c)

NLTK

d)

Statsmodels

20.

Which library is used for NLP?

a)

Pandas

b)

NLTK

c)

OpenCV

d)

PyTorch

21.

Accuracy is:

a)

(TP+TN)/(Total samples)

b)

TP/(TP+FP)

c)

TP/(TP+FN)

d)

2*(Precision*Recall)/(Precision+Recall)

22.

Precision focuses on:

a)

Correctly identifying all positives

b)

Correctly identifying all negatives

c)

Correctness of positive predictions

d)

Misclassification rate

23.

Recall is:

a)

TP/(TP+FN)

b)

TP/(TP+FP)

c)

(TP+TN)/(Total samples)

d)

FP/(FP+TN)

24.

F1-score balances:

a)

Accuracy & Recall

b)

Precision & Recall

c)

Accuracy & Precision

d)

Recall & Loss

25.

Which metric is best for imbalanced datasets?

a)

Accuracy

b)

F1-score

c)

RMSE

d)

26.

RMSE is always:

a)

Less than MSE

b)

Greater than or equal to MSE’s square root

c)

Negative

d)

Equal to R²

27.

R² (coefficient of determination) indicates:

a)

Correlation between predicted & actual values

b)

Error in predictions

c)

False positive rate

d)

Normalization factor

28.

A confusion matrix is used for:

a)

Regression

b)

Classification

c)

Clustering

d)

Dimensionality reduction

29.

ROC-AUC is useful for:

a)

Comparing regression models

b)

Evaluating classification thresholds

c)

Visualizing clustering

d)

Feature engineering

30.

Logloss measures:

a)

Error in classification probabilities

b)

Training speed

c)

Bias in regression

d)

Clustering variance

31.

Clustering metric ARI stands for:

a)

Adjusted Random Index

b)

Adjusted Rand Index

c)

Average Random Index

d)

Artificial Ranking Index

32.

Dimensionality reduction is useful for:

a)

Increasing dataset size

b)

Reducing computational complexity

c)

Labeling data

d)

Increasing accuracy only

33.

Which dimensionality reduction technique uses eigenvectors?

a)

PCA

b)

t-SNE

c)

K-means

d)

DBSCAN

34.

Which unsupervised algorithm detects anomalies?

a)

Logistic regression

b)

DBSCAN

c)

Naïve Bayes

d)

Linear regression

35.

Association rule learning is mainly used in:

a)

Customer segmentation

b)

Market basket analysis

c)

Regression

d)

Clustering

36.

CNNs are mainly used for:

a)

Sequential data

b)

Image processing

c)

Probability estimation

d)

Text summarization

37.

RNNs are best for:

a)

Image classification

b)

Sequential data like time series

c)

Regression

d)

Clustering

38.

Transformers replaced RNNs in NLP because:

a)

They require less memory

b)

They process sequences in parallel

c)

They cannot overfit

d)

They don’t use embeddings

39.

GANs are used for:

a)

Generating new synthetic data

b)

Feature scaling

c)

Regression

d)

Clustering

40.

Autoencoders are used for:

a)

Supervised learning

b)

Feature extraction & dimensionality reduction

c)

Classification only

d)

Data labeling

41.

A diffusion model is mainly used for:

a)

Data cleaning

b)

High-quality image generation

c)

Feature scaling

d)

Regression

42.

Overfitting occurs when:

a)

Model generalizes well

b)

Model memorizes training data

c)

Model under-learns

d)

Model is regularized

43.

Underfitting means:

a)

Model performs poorly on training data

b)

Model performs poorly on test data only

c)

Model has too many parameters

d)

Model generalizes too much

44.

A good fit is when:

a)

Training accuracy >> Test accuracy

b)

Training accuracy ≈ Test accuracy

c)

Training accuracy < Test accuracy

d)

None

45.

Deep reinforcement learning combines:

a)

Clustering & regression

b)

Deep learning & reinforcement learning

c)

CNNs & RNNs

d)

PCA & GANs

46.

ML focuses on:

a)

Manual feature extraction

b)

Automatic feature extraction

c)

Only image processing

d)

Only text

47.

DL focuses on:

a)

Manual feature engineering

b)

Automated feature extraction

c)

Only regression tasks

d)

Dimensionality reduction

48.

Which datatype is discrete?

a)

Temperature in Celsius

b)

Number of students in a class

c)

Height of a person

d)

Weight in kilograms

49.

Which preprocessing method is used for categorical variables?

a)

Standardization

b)

One-hot encoding

c)

Min-max scaling

d)

Normalization

50.

The ML pipeline generally ends with:

a)

Data collection

b)

Model deployment

c)

Feature engineering

d)

Hyperparameter tuning