Objective 2.2 Explain Data Collection/transformation Process in ML Workflow (transformations Include Standardization; Normalization; Log, Square-Root, and Logit Transformations) AIP-210 Practice Questions (Page 6)

Questions 26–27

1. 26application · medium

   A team is building a model to predict equipment failure. They have access to sensor data collected every second, maintenance logs updated monthly, and operator shift schedules. The data scientist notices that the sensor data has many missing values during network outages. Which data collection practice is most important to address before any transformation is applied?

   • AStandardize the sensor data to handle the missing values.
   • BCollect more sensor data to increase the sample size.
   • CApply a log transformation to the sensor data to reduce the impact of outliers.
   • DDocument the data collection process and record the reason for missing values.
   Reveal Solution Discussion Select an answer first
2. 27application · medium

   A data scientist is preparing features for a k-nearest neighbors (kNN) model. The dataset contains 'age' (range 18–90), 'annual_income' (range $20,000–$250,000), and 'credit_score' (range 300–850). The model will compute Euclidean distances between points. Which transformation approach should the data scientist apply to ensure no single feature dominates the distance calculation?

   • AApply a log transformation to each feature to compress the scale of the larger values.
   • BApply a square-root transformation to each feature to stabilize variance.
   • CApply standardization (z-score) to each feature so they all have mean 0 and standard deviation 1.
   • DApply min-max normalization to each feature to rescale them to the range [0, 1].
   Reveal Solution Discussion Select an answer first