Our AI Detects Your AI — Revealing the Secret Blockchain DApp World of Bots (Part 1 — EOS)

Image for post
Image for post

Abstract

Key Insights

Image for post
Image for post
Figure 1: Model Prediction Results For Bot Prevalence in Top 10 EOS DApps — Unique Accounts (left), Transaction Volume (right)
Image for post
Image for post
Figure 2: Bot vs. Human Accounts (left) and Transaction Activity (right)

Why does blockchain bot detection matter ?

Image for post
Image for post
Figure 3: Internet Bot Research Report — Distill Networks, 2019 [1]
Image for post
Image for post
Figure 4: Bitwise‘s report shows trading bots found in several crypto exchanges [2].

Background: Blockchain Ecosystem and DApp

Image for post
Image for post
Figure 5: Comparison of Top 3 DApp-Friendly Blockchains (EOS, TRON, ETH). April 2019 Dapp.Review — Full Version in Appendix
Image for post
Image for post
Figure 6: TokenInsight’s Q1 2019 DApp Report
Image for post
Image for post
Figure 7: DApp Hot 20 EOS Board. March 2019, TokenInsight.

Blockchain Bot vs. Human Behaviors

Image for post
Image for post
Figure 8: EOS DApp Bot Blockchain Transactions (addresses made anonymous)
Image for post
Image for post
Figure 9: Typical Blockchain DApp Bot Behavior
Image for post
Image for post
Figure 10: Typical Human Account Behavior

Four Blockchain Bot Behavior Sophistication Levels

Image for post
Image for post
Figure 11: Hyperactive bots (24x7, nonstop, relatively linear)
Image for post
Image for post
Figure 12 : Regular patterns — runs every 4 hours (left), runs every hour (right)
Image for post
Image for post
Figure 13: Daily Active Bots Employing Perturbation Techniques
Image for post
Image for post
Figure 14: Blockchain APT [3] hacker group identified in a blockchain DApp in Aug 2018. Each dot in this massive graph represents a blockchain address, and the center is the targeted DApp, Fomo3D.

Why are blockchain bots challenging to detect?

Bad Bot vs Good Bot in Blockchain World

What can we do as an industry?

How does AnChain.AI detect bots?

Image for post
Image for post
Figure 15: AnChain.AI convolutional neural network (CNN) model architecture
Image for post
Image for post
Figure 16: Deep learning model performance on training (blue) and validation (orange) set. Achieved 99%+ accuracy on the validation set.
Image for post
Image for post
Figure 17: Random forest machine learning model for bot detection. 99%+ ROC score.

Acknowledgements

About AnChain.AI

References:

Appendix:

Image for post
Image for post
Full Stats from DApp.Review website.

Blockchain data analytics firm providing security, risk, and compliance solutions.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store