Objective 7.1 Report Findings CERTIFIED-DATA-SCIENCE-PRACTITIONER Practice Questions (Page 9)

Questions 41–45

1. 41foundation · easy

   What is the purpose of post hoc explainable methods?

   • ATo increase the model's accuracy during training
   • BTo replace the model with a simpler one
   • CTo train a model that is inherently interpretable from the start
   • DTo explain a black-box model after it has been trained
   Reveal Solution Discussion Select an answer first
2. 42foundation · easy

   Which type of visualization is commonly used to show the impact of features on model predictions?

   • AHistogram of target variable
   • BPie chart
   • CLine chart of training loss
   • DSHAP summary plot
   Reveal Solution Discussion Select an answer first
4. 43application · medium

   A data scientist has trained a binary classification model and wants to explore how changing a specific feature value would alter the prediction for a particular instance. The goal is to understand the model's decision boundary interactively without retraining. Which tool is most suitable?

   • AA confusion matrix
   • BA static SHAP summary plot
   • CA learning curve
   • DGoogle's What-If Tool
   Reveal Solution Discussion Select an answer first
5. 44application · medium

   A credit union must comply with a new regulation requiring that loan decisions be fully explainable to customers. The compliance team is willing to accept a small reduction in predictive accuracy to guarantee transparency. The model will be retrained quarterly. Which modeling approach best aligns with this requirement?

   • AUse a decision tree with a maximum depth of 5 and document the splitting rules.
   • BUse a support vector machine with a linear kernel and report the weight vector.
   • CUse a deep neural network and apply LIME for each prediction.
   • DUse a random forest and rely on SHAP values for explanations.
   Reveal Solution Discussion Select an answer first
6. 45foundation · easy

   What is a key advantage of using Django over Flask for a web application that serves model predictions?

   • ADjango is a micro-framework that requires minimal setup
   • BDjango includes built-in features like an admin panel, authentication, and ORM
   • CDjango is specifically designed for machine learning model serving
   • DDjango is simpler to learn and has a smaller learning curve
   Reveal Solution Discussion Select an answer first