{"blocks": [{"key": "177acc43", "text": "Scenario", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "18b1a090", "text": "Evaluating a classifier that flags bad actors", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "5dcd0f20", "text": "Question", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "2508ab96", "text": "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?", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "63bb8d0b", "text": "Hints", "type": "header-two", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}, {"key": "10189aa4", "text": "Direct application of Bayes’ rule with symmetric 95 % true-positive and false-positive complements.", "type": "unstyled", "depth": 0, "inlineStyleRanges": [], "entityRanges": [], "data": {}}], "entityMap": {}}