Opportunity
The project explored how automated analysis could flag patterns associated with suspicious or bot-driven activity across blockchain networks for further investigation.
Challenge
- Market Manipulation: Users employing fake activity to manipulate crypto market dynamics.
- Large-Scale Analysis: The challenge of monitoring wallet activity across several blockchain networks.
- Real-Time Detection: Need for an efficient, automated system to promptly identify suspicious behaviors.
Solution
1D.works developed a sophisticated BigQuery-based machine learning system:
- Data-Driven Detection: Utilized reference datasets of known malicious activities to train machine learning models.
- Daily Monitoring: Automatically scans and analyzes wallet activities across blockchain platforms, identifying anomalies and suspicious behaviors.
- Multi-Blockchain Compatibility: Built data workflows for Ethereum, Tron, Polygon, and BNB.
Delivered Capabilities
- Scheduled Monitoring: Automated recurring analysis of wallet activity.
- Investigation Flags: Identified anomalous patterns for review rather than asserting malicious intent automatically.
- Cross-Blockchain Data Workflows: Applied a common analysis approach across Ethereum, Tron, Polygon, and BNB.
- Traceable Analysis: Produced structured outputs that could support follow-up investigation.
The system provided detection and review capabilities. Its effect on market manipulation, investor confidence, or market stability has not been claimed without supporting outcome data.
Explore our AI in Finance & Trading service. Read: Your Private Data Moat.