Objective 1.2 Analyze the Use Cases of ML Algorithms to Rank Them by Their Success Probability AIP-210 Practice Questions (Page 5)

Questions 21–25

1. 21foundation · medium

   Which factor most directly influences the success probability of an ML algorithm for a given use case?

   • AThe quality and quantity of the available data
   • BThe color scheme of the data visualizations
   • CThe number of lines of code in the algorithm implementation
   • DThe programming language used to deploy the model
   Reveal Solution Discussion Select an answer first
2. 22application · medium

   A city government wants to use ML to predict traffic congestion levels (low, medium, high) at various intersections. They have historical traffic sensor data, weather data, and event schedules. However, the data is only available for the past six months and contains many gaps. What is the MOST significant risk to the success of this ML project?

   • AThe deployment environment is not suitable for ML models
   • BThe business objective is not clearly defined
   • CThe problem is too complex for any ML algorithm
   • DThe data quantity and quality are insufficient to train a reliable model
   Reveal Solution Discussion Select an answer first
4. 23foundation · medium

   Which practical constraint is most likely to affect the choice between a deep learning model and a simpler model?

   • AThe color of the data
   • BThe number of stakeholders
   • CThe office location
   • DThe availability of large amounts of training data
   Reveal Solution Discussion Select an answer first
5. 24foundation · medium

   A bank wants to classify loan applications as 'approved' or 'denied' based on applicant data. Which type of algorithm is best suited?

   • APrincipal Component Analysis (PCA)
   • BK-Means
   • CLogistic Regression
   • DLinear Regression
   Reveal Solution Discussion Select an answer first
6. 25application · medium

   A healthcare startup is building a model to detect diabetic retinopathy from retinal images. They have 10,000 labeled images and a limited budget for GPU compute. The model will be deployed in clinics with standard hardware. Which algorithm should be ranked highest for success probability?

   • AA support vector machine (SVM) with an RBF kernel on raw pixel values
   • BA random forest trained on hand-crafted image features
   • CA pre-trained convolutional neural network (CNN) with transfer learning
   • DA logistic regression model on flattened pixel intensities
   Reveal Solution Discussion Select an answer first