NEW
Font size
WorksheetsSupervisedLearning_Regression
Total questions: 10
Worksheet time: 5mins
Supervised learning algorithms require:
Unlabeled data
Labeled input-output pairs
Only numerical data
Only categorical data
Which of the following is a supervised learning task?
Clustering customer groups
Predicting house prices
Discovering hidden patterns
Topic modeling
In linear regression, the line of best fit is chosen to:
Minimize the total number of data points
Maximize classification accuracy
Minimize the sum of squared errors
Maximize correlation only
The output of linear regression is typically:
A category
A continuous value
A distance metric
A cluster
The train-test split is mainly used to:
Reduce dataset size
Create labels
Evaluate model performance on unseen data
Improve visualization
Which term refers to the difference between predicted and actual values?
Activation
Entropy
Residual(Error)
Kernel
What does the test_size parameter in train_test_split define?
Number of features
Percentage of data used for testing
Model accuracy
Loss function
Which of the following is a commonly used error metric for regression?
Accuracy
Precision
Mean Squared Error (MSE)
Recall
Mean Squared Error (MSE) measures:
Average absolute difference between predictions and actual values
Square root of residuals
Average of squared differences between predictions and actual values
Maximum value in the dataset
If test_size = 0.2, how much data is used for training?
20%
40%
50%
80%
