Skip to content
SIH Buddyby Ganeev Singh
Dev
All problem statements
SIH26151Worth consideringacceptance 3/5

Dark web threat actor de-anonymization

National Technical Research Organisation (NTRO) · Blockchain & Cybersecurity · Software

The entity-graph and stylometry are solid, buildable work, but be honest that you are demonstrating a de-anonymisation method on constructed personas — you cannot manufacture the real adversary mistakes that genuine unmasking depends on.

What it actually is

Criminals operate on the dark web hidden behind Tor, which makes it very hard to work out who they really are. The ask is a system that gathers a threat actor's footprints across marketplaces and forums, links their handles, PGP keys and wallets into one profile, and tries to connect a hidden service or a rebranded persona back to a real-world entity.

What to build

An intelligence platform with three capabilities: an operational-security-failure detector that scans Tor hidden services for misconfigurations — exposed server-status pages, SSL certificates tied to clearnet domains, default banners, descriptor inconsistencies — and matches them against clearnet infrastructure to suggest likely origin servers; an entity-resolution graph linking handles, PGP keys, wallets and trust relationships across multiple marketplaces into a single actor profile; and an AI persona-linking layer using stylometry and behavioural profiling to connect rebranded or migrated personas to known actors, all queryable over a timeline through an analyst dashboard.

Smallest thing that wins the room

Feed the platform two marketplace personas that are actually the same actor and show it linking them through a shared PGP key and a stylometric match, then flag a hidden service whose SSL certificate leaks a clearnet domain, surfacing a candidate origin server.

How crowded this one gets

A guess, projected from the 2025 statements — the last year where both the submission counts and the winners were published.

Quiet70–160 teams expectedroughly 1 in 60–139 wins it

Quieter than 82% of the 226 · #41 of 226 by expected field

Few teams are likely to go here. The best odds on the board come from statements like this.

Why: defence, intelligence and space bodies drew small fields.

This is a guess, not a fact

Nobody has published 2026’s numbers yet. This is an analysed estimate from last year’s pattern, so please do not take it as the truth — check the live counter on the SIH portal before you decide anything. The range covers the middle half of likely outcomes, so one statement in two lands outside it. Entry closes at 500 ideas per statement, so no range goes past that — a statement that reaches the cap fills and shuts rather than drawing an unlimited crowd. The model reads only three things a team can see before choosing — software or hardware, the theme, and what kind of body posted it — and those explain about a quarter of the variation in last year’s field sizes (R² 0.25 on held-out statements). Trust the band more than the number, and the ordering more than either. It cannot see how good your idea is, which is the part that actually decides it.

The scores

The number is the shorthand. The line under it is the reason.

What you will be writing

  • Stylometry / authorship attribution (BERT embeddings)
  • PGP key and wallet entity-resolution graph (NetworkX)
  • Tor hidden-service descriptor + certificate analysis
  • Cross-marketplace handle linkage
  • Behavioural profiling from posting patterns
  • Timeline-queryable analyst dashboard
  • Threat intelligence
  • OSINT attribution
  • Dark web forensics

Prior art to read before you start

cross-marketplace actor de-anonymisation · stylometric persona linking · hidden service misconfiguration attribution

Analysed by Claude Opus. Every score above is a judgment call with its reasoning attached — kindly cross-check this against the official statement on the SIH portal before your team commits to it.