How Data, Algorithms, and Models Form AI Foundations AIE Practice Questions (Page 5)

Questions 21–25

1. 21application · medium

   A retail company wants to build a model that predicts which customers are likely to churn. They have collected historical purchase records, customer service interactions, and demographic data. The data contains many missing values and inconsistent formats. What should the team do first to ensure the model can be effectively trained?

   • ATrain the model immediately to see what patterns emerge
   • BClean and preprocess the data to handle missing values and standardize formats
   • CSelect a complex algorithm that can handle messy data automatically
   • DCollect more data to overcome the quality issues
   Reveal Solution Discussion Select an answer first
2. 22foundation · easy

   Which of the following best describes how an algorithm processes data to perform a task?

   • AIt randomly selects data points to create a representative sample.
   • BIt applies a sequence of defined operations to the data to produce an output.
   • CIt converts data into a visual representation for human interpretation.
   • DIt stores the data in a structured format for later retrieval.
   Reveal Solution Discussion Select an answer first
4. 23application · medium

   A weather forecasting agency wants to predict daily temperature. They have 10 years of historical weather data. They train a model using a neural network. After training, they evaluate the model on a test set and find it performs well. What does this indicate?

   • AThe model has memorized the training data perfectly
   • BThe model has learned general patterns from the data that can be applied to new weather conditions
   • CThe neural network algorithm is the only suitable choice for this task
   • DThe model will always be accurate for any future weather
   Reveal Solution Discussion Select an answer first
5. 24expert · hard

   A financial firm wants to build a model to detect insider trading. They have a small dataset of confirmed cases and a large dataset of normal trades. They need to minimize false positives to avoid investigating innocent employees. Which algorithm choice best addresses this constraint?

   • AUse an ensemble method with class weighting to penalize false positives more heavily
   • BUse a simple decision tree because it is easy to interpret
   • CUse a random forest without adjusting class weights
   • DUse a neural network with many layers to capture complex patterns
   Reveal Solution Discussion Select an answer first
6. 25expert · hard

   A city government wants to use AI to allocate resources for emergency response. They have historical incident data, but the data is biased because certain neighborhoods have historically received more attention. They want to ensure the model does not perpetuate this bias. What is the most important step?

   • ARely on the model's predictions without human oversight
   • BUse a more complex algorithm to automatically correct for bias
   • CCollect more data from the over-served neighborhoods to balance the dataset
   • DAudit the training data for bias and adjust the algorithm or data to mitigate it
   Reveal Solution Discussion Select an answer first