Detect and Reduce Spammy Friend Requests Effectively

Q: Detect and Reduce Spammy Friend Requests Effectively

This question evaluates applied machine learning system design and data science competencies, including discriminative feature selection, classification modeling, label bootstrapping strategies, evaluation metrics, deployment constraints, product integration, and privacy and fairness considerations.

Q: How do I approach Machine Learning interview questions?

Machine Learning questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master machine learning interviews.

Question

Scenario

You are designing an ML system to detect and reduce spammy friend requests on a social-network platform. The system should run near real time at the moment a request is sent and integrate with product surfaces (e.g., inbox ranking, rate limits, warnings).

Questions

Definition and Features: How would you define a "spammy" friend request and choose discriminative features?
Modeling: How would you build a classification model (data, validation, algorithms, metrics, deployment constraints)?
No Labels: If no labeled data exist, what strategies would you use to bootstrap training data and models?
Product Integration: After deployment, how can the model be leveraged to improve user experience?
Thresholding: How would you balance precision and recall and decide on the operating threshold?

Assume you can log request-level events (sender, recipient, time, device/IP, graph context) and respect privacy and fairness constraints. If needed, make minimal reasonable assumptions and call them out.

Detect and Reduce Spammy Friend Requests Effectively

Scenario

Questions

Solution

Comments (0)

Detect and Reduce Spammy Friend Requests Effectively

Detecting Spammy Friend Requests on a Social Network

Scenario

Questions

Solution

Comments (0)