Question 694 of 753
AI Associate Data for AI Practice Question
A data architect is designing a data model for Einstein Discovery. The data includes categorical variables with high cardinality (e.g., postal codes). What is the best practice to handle such features?
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
Group them into higher-level categories (e.g., region).
Grouping high-cardinality categories into broader categories reduces overfitting and improves model stability.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✗
Encode them as one-hot vectors.
Why it's wrong here
One-hot encoding with high cardinality creates too many features.
- ✗
Exclude them from the model.
Why it's wrong here
Excluding may lose valuable information.
- ✗
Use the raw values without transformation.
Why it's wrong here
Raw values with many levels can cause overfitting.
- ✓
Group them into higher-level categories (e.g., region).
Why this is correct
Reduces cardinality while preserving signal.
About these practice questions
Courseiva creates original exam-style practice questions with explanations and wrong-answer analysis. It does not publish real exam questions, exam dumps, or protected exam content. Learn why practice questions differ from exam dumps →
Same concept, more angles
2 more ways this is tested on AI Associate
These questions test the same concept from different angles. Work through them to make sure you can recognise it however the exam phrases it.
Variation 1. During data transformation, a data scientist applies one-hot encoding to a categorical feature with 50 unique values. The resulting dataset has 50 new columns. What is a potential drawback of this transformation?
medium- A.Reduction in training time
- B.Increased interpretability of the model
- ✓ C.High cardinality leading to sparse data and overfitting
- D.Loss of ordinal information in categories
Why C: One-hot encoding a categorical feature with 50 unique values creates 50 binary columns, each representing one category. This high cardinality leads to a very sparse matrix (most entries are 0), which can cause the model to overfit by learning noise from rare categories, especially when the dataset is not large enough to support such dimensionality.
Variation 2. Which data transformation is most appropriate for converting categorical variables into numerical format for a machine learning model?
easy- A.Normalization.
- ✓ B.One-hot encoding.
- C.Principal component analysis.
- D.Standardization.
Why B: One-hot encoding is the correct transformation because it converts categorical variables into a binary vector representation, where each category becomes a separate column with a 1 or 0. This allows machine learning models to interpret categorical data without implying any ordinal relationship, which is essential for algorithms that rely on numerical distances or linear algebra.
Last reviewed: Jun 23, 2026
This AI Associate practice question is part of Courseiva's free Salesforce certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the AI Associate exam.
Question Discussion
Share a tip, memory trick, or ask about the reasoning behind this question. Do not post real exam questions, leaked content, braindumps, or copyrighted exam material. Comments are moderated and may be removed without notice.
Sign in to join the discussion.