DA0-002 Data Analysis Practice Question

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Increasing the maximum depth of the tree

  Why it's wrong here

  Increasing maximum depth allows the tree to grow deeper, fitting training data more closely and exacerbating overfitting. A deeper tree can memorize noise and outliers, reducing performance on unseen data. This is the opposite of what the analyst needs to prevent overfitting in churn prediction.

• ✗

  Setting the minimum samples per leaf to 1

  Why it's wrong here

  Setting minimum samples per leaf to 1 allows the tree to create leaves with single observations, which perfectly fits training data and causes severe overfitting. A larger minimum leaf size constrains the tree, promoting generalization. For churn prediction, a value greater than 1 is typically used to avoid overfitting.

• ✓

  Pruning the tree after full growth

  Why this is correct

  Pruning involves growing a full tree and then removing branches that provide little predictive power, reducing complexity and overfitting. This technique, such as cost-complexity pruning, balances bias and variance, improving generalization. For churn classification, pruning helps avoid capturing noise in the training data, leading to more robust predictions on new customers.

• ✗

  Using all available features without selection

  Why it's wrong here

  Including all features, especially irrelevant ones, can lead to a more complex tree that overfits. Feature selection or dimensionality reduction helps simplify the model. Without selection, the tree may split on noise, harming generalization. This approach does not address overfitting and could worsen it.