Objective 1.2 Analyze the Use Cases of ML Algorithms to Rank Them by Their Success Probability AIP-210 Practice Questions (Page 2)

Questions 6–10

1. 6application · medium

   A small e-commerce company wants to build a churn prediction model. They have a dataset of 10,000 customers with 50 features. The team has limited ML experience and needs to deploy the model quickly. Which algorithm should be ranked highest for success probability?

   • AA random forest with 1,000 trees
   • BA logistic regression model
   • CA deep neural network
   • DA support vector machine with RBF kernel
   Reveal Solution Discussion Select an answer first
2. 7application · medium

   A fraud detection team is comparing two models on the same dataset. Model A has an accuracy of 99% but a recall of 0.20 for the fraud class. Model B has an accuracy of 97% but a recall of 0.85 for the fraud class. The cost of missing a fraud case is very high. Which model should be ranked higher for success probability?

   • AModel A, because higher accuracy is always better
   • BModel B, because higher recall means fewer missed fraud cases
   • CModel A, because it has higher precision
   • DBoth models are equally good because accuracy is similar
   Reveal Solution Discussion Select an answer first
4. 8application · medium

   A logistics company wants to group its delivery routes into clusters to optimize resource allocation. They have GPS coordinates and delivery times for thousands of routes, but no predefined labels. Which type of ML problem is this, and which algorithm is most appropriate?

   • AClustering; use K-means
   • BRegression; use linear regression
   • CDimensionality reduction; use PCA
   • DClassification; use a decision tree
   Reveal Solution Discussion Select an answer first
5. 9application · medium

   A media company wants to classify news articles into topics (sports, politics, tech, etc.). They have a large labeled corpus of 1 million articles. The model will be deployed in a real-time API with low latency requirements. Which algorithm should be ranked highest for success probability?

   • AA random forest on raw word counts
   • BA K-nearest neighbors classifier
   • CA linear SVM with TF-IDF features
   • DA deep learning model like BERT
   Reveal Solution Discussion Select an answer first
6. 10foundation · medium

   When comparing two regression algorithms, which metric is commonly used to evaluate their performance?

   • APrecision
   • BF1-score
   • CMean Absolute Error (MAE)
   • DRecall
   Reveal Solution Discussion Select an answer first