Core Platform Team/Initiatives/Sockpuppet Detection API

Initiative Vision

< Initiatives

Vision:
  • Use machine learning to detect multiple accounts controlled by the same person used for sockpuppetry
Target Group(s):
  • Wiki moderators
  • Anti-Harassment Tools
  • Research
  • Machine Learning
Needs:
  • (Moderators) Check an account to see if it shows signs of being a sockpuppet
  • (AHT) Provide a user interface for checking accounts for sockpuppetry
  • (Research) Host the machine learning model they created for detecting sockpuppets
  • (Machine Learning) API interface to machine learning model
  • (Machine Learning) Interim model hosting while ML platform is being built
Product:
  • Kubernetes image running MLFlow to host the ML Model
  • API facade for MLFlow, only allowing access to checkuser users
  • Deployment tools for updating and revising the model
Aligned Goals:
  • "Software platforms with integrated machine learning, rich media, and structured data components, and associated tooling for internal and external development and reuse of code and content."



Subpages