Skip to content

We use machine learning and graph algorithms to analyze the attributes of TRON addresses with the goal of assisting in the tracking of illicit funds.

License

Notifications You must be signed in to change notification settings

slowmist/automatic-tron-address-clustering

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Automatic TRON Address Clustering

Introduction

In this age of information boom, the SlowMist security team discovered that malicious actors frequently utilize social media, phishing websites, and other methods to steal users' digital assets. Our address identification mechanism primarily uses graph and machine learning algorithms to evaluate the public transaction history of TRON's users and label their addresses into 3 separate categories: Hot wallet, Cold wallet, and Common users. This is accomplished by combining the graph algorithm with node mining techniques to execute large-scale mining of transaction attributes of user address transactions. This also has the ability to perform influence analysis, community discovery based on tag propagation algorithm, and transaction path discovery based on graph traversal algorithm.

Architecture

avatar

Deployment Environment

Node Server

  • Only One
  • CPU:16 cores
  • RAM:32G
  • SSD:1.5T+
  • Bandwidth:100M
  • Docker Server:20.10.5
  • Deploy the TRON Node and Kafka

Algorithm Server

  • At Least One
  • CPU:16 cores
  • RAM:32G
  • SSD:500G+
  • Bandwidth:100M
  • Docker Server:20.10.5
  • Deploy Spark

Platform Server

  • At Least One
  • CPU:16 cores
  • RAM:32G
  • SSD:500G+
  • Bandwidth:100M
  • Docker Server: 20.10.5
  • Deploy graph database: Nebula Graph

Install dependency component

Install Reference

Core Code and Usage

Machine Learning

The machine learning component primarily employs the ensemble learning model, with label data derived from publicly available information on the TRON explorer, mixed with manual data analysis to organize and train the model.

avatar

More Detail

Graph Algorithms

We generate the point and edge data from the data set and import it into the graph database for use in the next graph query and graph algorithm calculation.

avatar

More Detail

Summarize

Our goal this time is to present an advanced system utilizing machine learning and graph computing to analyze attributes of TRON addresses to identify common user addresses, hot wallet addresses, and cold wallet addresses in the TRON ecosystem. At the same time, our analysis will examine each address's characteristics and relationships to aid in the tracking of hacker addresses and activity, allowing relevant personnel to access additional insight into a wallet.

Future

  1. Improve the algorithm to enhance recognition accuracy
  2. Access additional TRON ecosystem functionality, expand data scope, and enhance processing efficacy.
  3. Increase application scope and generate value for more users.
  4. To better serve the TRON community and its users, we will continuously enhance MistTrack's user experience and capabilities. Additional servers will be added to increase performance and introduce new algorithms to assist the needs of various users. Such as implementing the Graph Neural Networks algorithm (using the DGL component) for community research in the near future. With each new update, MistTrack will be able to provide additional benefits to our users.

Links

  1. Our submission on TRON Grand Hackathon Season 3

  2. Our submission on TRON Grand Hackathon 2022

  3. MistTrack Platform Intro

  4. MistTrack Dashboard

About

We use machine learning and graph algorithms to analyze the attributes of TRON addresses with the goal of assisting in the tracking of illicit funds.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published