Identify Metrics to Detect Fake-Account Activity on Facebook

Q: Identify Metrics to Detect Fake-Account Activity on Facebook

This question evaluates a Data Scientist's competency in statistical inference, anomaly detection metrics, behavioral feature engineering, and classifier trade-off analysis.

Q: How do I approach Statistics & Math interview questions?

Statistics & Math questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master statistics & math interviews.

Question

Detecting and Measuring Fake Accounts

Scenario

Facebook wants to understand and curb fake-account activity both over the past month and at sign-up time.

Tasks

Month-over-month proxies: List product or engagement metrics that could indicate changes in fake-account activity.
Post-registration signals: Identify behavioral signals that help distinguish fake from real accounts after signup.
Registration-time detection: Can we flag a fake account during the registration flow? Propose features and quick checks.
Bayesian update: Given a prior fake-account rate of 2% and observing 18 flags in a cohort of 600 new accounts, compute the posterior probability that the next account is fake. Use Beta–Binomial conjugacy and show steps.
Classifier objective: For evaluating a fake-account classifier, would you optimize precision or recall? Discuss the trade-offs and wider impact on Facebook.

Identify Metrics to Detect Fake-Account Activity on Facebook

Detecting and Measuring Fake Accounts

Scenario

Tasks

Solution

Comments (0)

Identify Metrics to Detect Fake-Account Activity on Facebook

Overview

Detecting and Measuring Fake Accounts

Scenario

Tasks

Solution

Comments (0)