Reputation System for Financial Markets

[akolonin]

Reputation System for Financial Markets (might be interesting for SDAO and SNET)

• Explore the organic hodling vs. manipulative pumping-dumping patterns on the market to generate market signals
• Explore reputations of the market agents exposing these patterns
• Assess the reliability of the generated market signals from perspective of reputations of the agents generating them
• Come up with framework of market agent reputation assessment along with market signal detection and filtering
• Run the experiments according to the above on
  — simulation model
  — Ethereum Data (will need at least 8GB RAM to run the experiments)
• Have the above implemented in either
  — a) https://github.com/aigents/aigents-java/blob/master/src/main/java/net/webstructor/peer/Reputationer.java (can have full @akolonin support, need to learn Java, no simulation present)
  — b) https://github.com/singnet/reputation (all in Python, simulation prototype is in-place but might need to get changed)
  — c) combine both (RS engine from a, simulation from b)
  Resources:
  https://blog.singularitynet.io/reputation-assessment-vs-reputation-management-d3a71445505a
  https://blog.singularitynet.io/minimizing-recommendation-fraud-7dabbee8fc00
  https://blog.singularitynet.io/utilizing-singularitynets-reputation-system-to-prevent-scams-5c7da7020bc9