Estimate Fake Accounts Using Data Signals and Sampling
Quick Overview
This interview question evaluates metric design, causal reasoning, experiment setup, diagnostics, SQL/statistical checks, and recommendations in a realistic interview setting. A strong answer for Estimate Fake Accounts Using Data Signals and Sampling states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Estimate Fake Accounts Using Data Signals and Sampling
Company: Meta
Role: Data Scientist
Category: Analytics & Experimentation
Difficulty: hard
Interview Round: Onsite
##### Scenario
Social media company wants to estimate the number of fake accounts on the platform.
##### Question
How would you design an analytics approach to estimate the proportion or absolute count of fake accounts on Facebook? What data signals, sampling strategy, and validation method would you use?
##### Hints
Consider random sampling, supervised classification, manual labeling, capture-recapture, and confidence intervals.
Quick Answer: This interview question evaluates metric design, causal reasoning, experiment setup, diagnostics, SQL/statistical checks, and recommendations in a realistic interview setting. A strong answer for Estimate Fake Accounts Using Data Signals and Sampling states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.
Estimate Fake Accounts Using Data Signals and Sampling
Estimating Fake Accounts on a Social Network
Background
A large social platform wants to estimate the proportion and absolute count of fake accounts on the service. "Fake" includes accounts that are inauthentic (bots, impersonations, coordinated inauthentic behavior), excluding clearly legitimate users. Estimates should be produced for a defined time window (e.g., monthly) and a defined population (e.g., all accounts, or active accounts in the last 30 days).
Task
Design an analytics approach to estimate the prevalence (percentage) and count of fake accounts on Facebook. Specify:
Data signals/features you would use.
A sampling strategy for labeling and estimation.
A modeling approach to classify fakes and estimate prevalence.
A validation plan and how you would compute confidence intervals.
Hints: Consider random/stratified sampling, supervised classification with manual labeling, capture–recapture, and confidence intervals.
Constraints & Assumptions
Preserve the scope, facts, inputs, and requested outputs from the prompt above.
If the prompt leaves a detail unspecified, state a reasonable assumption before relying on it.
Keep the answer interview-ready: concise enough to present, but concrete enough to implement or evaluate.
Clarifying Questions to Ask Guidance
Clarify the business objective, unit of analysis, time window, exposure definition, and primary metric.
State assumptions about instrumentation, randomization, sample size, and data quality.
Separate descriptive analysis from causal claims.
What a Strong Answer Covers Guidance
A metric framework with primary, guardrail, and diagnostic metrics.
A credible analysis or experiment design with clear assumptions and bias checks.
SQL/statistical logic for segmentation, variance, confidence, and data validation where relevant.
An actionable recommendation that explains trade-offs and next steps.
Follow-up Questions Guidance
What sanity checks would you run before trusting the result?
How would you handle novelty effects, seasonality, or selection bias?