Machine Learning Quiz

Which algorithm(s) is/are most suitable for predicting house prices based on numerical features such as square footage, number of bedrooms, and location?

• K-Means

• Linear Regression

• Decision Tree Regressor

• Logistic Regression

Which of the following transformations can help when data has a right-skewed distribution?

• Min-Max Scaling

• Log Transformation

• Standardization (Z-score scaling)

• Box-Cox Transformation

Which of the following techniques can be used to handle imbalanced datasets?

• Oversampling the minority class

• Undersampling the majority class

• Using a different loss function such as weighted cross-entropy

• Applying PCA to the dataset

Which of the following is true about Logistic Regression?

• It is a linear model used for classification

• It outputs a probability score between 0 and 1

• It works only for binary classification

• It assumes independence of predictor variables

What does Log Loss measure in a classification problem?

• The difference between predicted and actual values

• The error in regression tasks

• The log of the absolute difference between actual and predicted values

• The difference between predicted and actual probabilities

Why might Decision Trees overfit the data?

• They make strong linear assumptions

• They can grow too deep and memorize training data

• They are sensitive to changes in the training dataset

• They use distance-based similarity measures

Which of the following scenarios would be best suited for K-Means clustering?

• Identifying customer segments based on purchasing behavior

• Predicting house prices based on historical data

• Classifying emails as spam or not spam

• Finding natural groupings in unlabeled customer demographic data

A classification model shows high accuracy on training data but poor performance on test data. Which of these techniques could help? (Select all that apply)

• Increase model complexity

• Add regularization

• Collect more training data

• Feature selection

What metric is best for evaluating an imbalanced classification problem?

• Accuracy

• Precision

• Recall

• F1-score

When should you apply log transformation to your features?

• When the data is normally distributed

• When the data is highly skewed

• When dealing with categorical variables

• When features have negative values