MAC Predictive Modeling 2 — Questions and Answers
Question 1: Which metric measures the proportion of actual positives correctly identified by a classification model?
- Precision
- Recall (Correct answer)
- Specificity
- F1 Score
Correct answer: Recall
Recall (also called sensitivity or true positive rate) measures the proportion of actual positives that the model correctly identifies.
Question 2: In the context of predictive modeling, what does 'overfitting' mean?
- The model performs poorly on both training and test data
- The model learns noise in the training data and performs poorly on new data (Correct answer)
- The model uses too few features to make predictions
- The model takes too long to train on large datasets
Correct answer: The model learns noise in the training data and performs poorly on new data
Overfitting occurs when a model memorizes training data noise, resulting in high training accuracy but poor generalization to unseen data.
Question 3: A marketing team uses RFM analysis. What does 'F' stand for?
- Frequency (Correct answer)
- Forecast
- Funnel
- Feature
Correct answer: Frequency
In RFM analysis, F stands for Frequency — how often a customer makes a purchase within a given time period.
Question 4: Which ensemble method trains multiple decision trees on random subsets of data and averages their predictions?
- Gradient Boosting
- AdaBoost
- Random Forest (Correct answer)
- Bagging with Logistic Regression
Correct answer: Random Forest
Random Forest builds many decision trees on bootstrapped data samples with random feature subsets and averages predictions to reduce variance.
Question 5: What is the purpose of a confusion matrix in evaluating a predictive model?
- To visualize the distribution of input features
- To show true positives, false positives, true negatives, and false negatives (Correct answer)
- To measure the correlation between predictor variables
- To display the model's learning curve over training epochs
Correct answer: To show true positives, false positives, true negatives, and false negatives
A confusion matrix summarizes classification results by showing counts of true positives, false positives, true negatives, and false negatives.
Question 6: Which technique is used to reduce dimensionality by transforming features into uncorrelated principal components?
- K-Means Clustering
- Principal Component Analysis (PCA) (Correct answer)
- LASSO Regression
- Gradient Descent
Correct answer: Principal Component Analysis (PCA)
PCA projects data onto orthogonal principal components that capture the most variance, reducing the number of features while preserving information.
Question 7: In a customer churn prediction model, a high false negative rate means the model is:
- Flagging too many loyal customers as churners
- Missing many customers who will actually churn (Correct answer)
- Correctly identifying all churners
- Overestimating the churn rate across all segments
Correct answer: Missing many customers who will actually churn
A high false negative rate means the model fails to predict churn for many customers who will actually leave, leading to missed retention opportunities.
MAC Predictive Modeling 2
Which metric measures the proportion of actual positives correctly identified by a classification model?