{"blocks":[{"key":"382bl","text":"Context & Assumptions","type":"header-two","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"5aubl","text":"On a large social network (similar to Facebook), only 1% of accounts are malicious or \"bad\"","type":"unordered-list-item","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"1s5gn","text":"Bad accounts send friend requests at a rate ten times higher than good accounts","type":"unordered-list-item","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"b6pk5","text":"You have developed a classification model that achieves a 95% true positive rate (TPR) and a 95% true negative rate (TNR)","type":"unordered-list-item","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"2eftf","text":"Questions","type":"header-two","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"d88ec","text":"1. Single Friend Request Probability","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"7ul9f","text":"Estimate the probability that a received friend request comes from a bad account.","type":"unstyled","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"45mg5","text":"2. Multiple Friend Requests","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"9s79l","text":"Determine the likelihood that, out of five friend requests, at least one originates from a bad account.","type":"unstyled","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"b03ju","text":"3. Model Reliability","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"58tus","text":"If the model flags an account as bad, how likely is it to be truly malicious, given the stated TPR and TNR?","type":"unstyled","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"9vf2a","text":"4. Data & Features","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"f1uig","text":"Identify which types of data (e.g., behavior logs, friend-request patterns, reported incidents) would be most relevant to classify accounts accurately.","type":"unstyled","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"9qh0p","text":"5. Assessing \"Bad Account\" Prevalence","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"a2c5h","text":"Propose methods (such as stratified or random sampling) to determine whether the bad-account issue is substantial enough to warrant further intervention.","type":"unstyled","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"felu1","text":"6. Defining a \"Bad User\"","type":"header-three","depth":0,"inlineStyleRanges":[],"entityRanges":[],"data":{}},{"key":"6tn6q","text":"Outline what characteristics or behaviors would qualify an account as \"bad\" (e.g., spammer, scammer, bot).","type":"unstyled","depth":0,"in