Determine Posterior Probability of Bad User Prediction
Company: Meta
Role: Data Scientist
Category: Statistics & Math
Difficulty: easy
Interview Round: Onsite
##### Scenario
Evaluating a classifier that flags bad actors
##### Question
5 % of users are actually bad. The model labels a user correctly with 95 % accuracy for both classes. If the model predicts a user is bad, what is the posterior probability the user is truly bad?
##### Hints
Direct application of Bayes’ rule with symmetric 95 % true-positive and false-positive complements.
Quick Answer: This question evaluates a candidate's understanding of Bayesian reasoning and conditional probability, focusing on concepts such as prevalence, sensitivity, specificity, and posterior probability in classifier evaluation within the Statistics & Math domain.