Explain Type I/II errors vs precision/recall
Company: TikTok
Role: Data Scientist
Category: Statistics & Math
Difficulty: easy
Interview Round: Technical Screen
## Questions
1. Define **Type I error** and **Type II error** in hypothesis testing, and map them to **false positives** and **false negatives**.
2. Explain how Type I/II errors relate (or don’t directly relate) to **precision** and **recall** in binary classification.
3. Give concrete examples of when **Type I error is worse** vs when **Type II error is worse**.
4. Given a scenario (e.g., medical test, fraud detection, spam filter, moderation), identify which outcome is Type I vs Type II and which metric you would prioritize.
Quick Answer: This question evaluates understanding of statistical hypothesis testing (Type I and Type II errors) and their relationship to binary classification metrics such as precision and recall, situated in the Statistics & Math domain and the binary classification/machine learning category, testing both conceptual understanding and practical application.