Identifying Family Connections

Requirements Clarification & Assessment

1. Objective Understanding:

   • Determine the primary goal: Is the algorithm for user recommendations, social tree mapping, or advertising?
   • Understand the level of accuracy required for the algorithm to be deemed successful.
   • Clarify the acceptable balance between precision and recall based on the use case.

2. Data Availability:

   • Identify available datasets: User profiles, photos, posts, and connections.
   • Determine if labeled data (confirmed sibling relationships) is available for supervised learning.
   • Assess the quality and completeness of the data.

3. Feature Identification:

   • Common features: Age, surname, hometown, mutual friends, tagged photos.
   • Derived features: Interaction frequency, shared events, and location overlap.

4. Stakeholder Expectations:

   • Understand the expectations of product managers and engineers regarding implementation timelines and resource allocation.
   • Discuss how the algorithm will be integrated into existing systems.

5. Ethical Considerations:

   • Address privacy concerns related to using personal data for sibling identification.
   • Ensure compliance with data protection regulations.