AI-Powered Monitoring & Analysis of Bitcoin Transaction Traffic
“This project involves building a smart detective tool that tracks suspicious Bitcoin activity by linking internet connection data with blockchain transactions. It helps investigators identify criminals who use digital currency to hide illegal money transfers by spotting patterns that look like money laundering.”
Deconstructed Problem Blueprint
WHAT TO BUILD?
An offline-capable forensic analytics platform that ingests network metadata and blockchain transaction logs to perform graph-based anomaly detection and entity clustering.
WHO IS IT FOR?
National Technical Research Organisation (NTRO) analysts, cyber-crime investigators, and financial intelligence units.
WHY DOES IT EXIST?
Criminals exploit the anonymity of Bitcoin to move illicit funds. Current tools often struggle to correlate the physical network layer (IP addresses) with the digital ledger layer (wallets), leaving a gap in tracking the 'cashing out' process.
INPUT → PROCESS → OUTPUT ARCHITECTURE PIPELINE
CSV/JSON/XML files containing timestamps, source/destination IPs, ports, TXIDs, wallet addresses, amounts, and fees.
Graph construction (NetworkX/Neo4j), feature engineering for transaction behavior, unsupervised clustering (DBSCAN/K-Means) for entity grouping, and supervised anomaly detection (Isolation Forests/XGBoost).
A visual dashboard displaying a ranked list of suspicious entities, confidence scores, and an explainable 'reasoning' log for each flagged transaction.
CONSTRAINTS
- Must be a fully offline solution (no external API calls during analysis).
- Must run on a standard Linux environment.
- Must handle bulk data ingestion efficiently without crashing.
- Explainability is mandatory (cannot be a black-box model).
JURY METRICS
- Precision and Recall of anomaly detection.
- Graph traversal speed for large datasets.
- Clarity and intuitiveness of the visualization dashboard.
- Quality of the 'Explainability' report (how well it justifies a flag).
CORE BOTTLENECK
The 'Correlation' challenge: Linking ephemeral IP addresses to persistent blockchain wallets when multiple users might share the same IP or use VPNs/Tor.
Why You Should Choose This PS (Pros)
- High national security impact.
- Clear, well-defined technical scope.
- Strong potential for integration with existing government forensic toolkits.
- Excellent opportunity to showcase advanced graph theory and ML skills.
Challenges & Risks to Watch (Cons)
- Requires deep understanding of blockchain mechanics.
- Risk of 'over-engineering' the UI at the expense of the core ML engine.
- Synthetic data generation needs to be highly realistic to be credible.
Recommended Tech Stack & Grand Finale Features
Killer Features to Impress the SIH Jury
View Official SIH 2026 Portal Raw Text▼
The objective of problem statement is to design and build a complete system (offline) that ingests bulk Bitcoin transaction/network metadata (in CSV/JSON/XML), correlates network-layer (IP/port/timing) observations with blockchain-layer (wallet/TXID/amount) data, and applies AI/ML to detect anomalies, cluster entities, and generate prioritized, explainable investigative leads.
ii.Suggested AI/ML Focus Areas Attach Table Here of AI/ML Focus Areas iii.Dataset: Parameters & Synthetic Generation Participants will work with a synthetic dataset modelled on real Bitcoin P2P/transaction fields (no real seized or live-intercept data will be provided). Minimum fields: timestamp, src_ip, dst_ip, src_port, dst_port, txid, input_addresses[], output_addresses[], input_amounts[], output_amounts[], geo_country/asn (integrate open source downloadable Geo IP database).