Strengthening Decentralized Identity with Cross-Platform Behavioral Proofs
Social Entropy Layered DIDs (SEL-DIDs) is a novel framework that enhances Decentralized Identifiers (DIDs) with cross-platform behavioral and social signals to solve three critical problems in decentralized identity:
- Sybil Resistance - Prevent fake identity attacks without centralized KYC
- Account Recovery - Enable key recovery without custodial services
- Trust Bootstrapping - Support passwordless authentication with human verification
Unlike systems that derive keys from social data (which is insecure), SEL-DIDs treat behavioral evidence as layered, revocable credentials attached to a cryptographic root identity.
┌─────────────────────────────────────────────────────────────────┐
│ SEL-DID Architecture │
├─────────────────────────────────────────────────────────────────┤
│ Web2 Platforms ──► Data Vault (DWN) ──► SEV Derivation ──► RP │
│ + PoP Systems (User-controlled) + ZK Prover │
│ ▲ ▲ │
│ │ │ │
│ DID + Keys ────────────────┘ │
│ (sk, pk) │
└─────────────────────────────────────────────────────────────────┘
- Cryptographic Root - Standard DID with high-entropy keypair (did:key, did:ion, etc.)
- Social Entropy Vector (SEV) - Aggregated behavioral features from multiple platforms
- Verifiable Credentials - Platform-issued attestations stored in user-controlled vaults
- Zero-Knowledge Proofs - Privacy-preserving verification of SEV properties
SEL-DIDs increase the cost of forging identities by requiring long-lived, cross-platform behavioral consistency that cannot be cheaply replicated at scale.
Relying parties verify that a policy is satisfied (e.g., "identity has 3+ years of consistent activity") without seeing raw behavioral data, contact graphs, or platform accounts.
All credentials and SEV computations remain in user-controlled data vaults (Decentralized Web Nodes), never in centralized databases.
No single platform is required. The system degrades gracefully when platforms are unavailable or uncooperative.
| System | Evidence Type | Centralization | Privacy Model |
|---|---|---|---|
| BrightID | Social graph + verification parties | Single protocol/foundation | Pseudonymous graph proofs |
| Gitcoin Passport | Multi-stamp credentials | Gitcoin-operated scoring | Stamps partly on-chain |
| Worldcoin | Biometric (iris) | Centralized hardware/registry | Controversial biometric handling |
| SEL-DID | Cross-platform behavioral SEV | No mandatory operator | ZK proofs over user-controlled data |
- Decentralized Governance - One-human-one-vote without KYC
- Quadratic Funding - Sybil-resistant public goods funding
- Civic Platforms - Local voting with location + social verification
- Reputation Systems - Portable, privacy-preserving reputation
- Passwordless Authentication - Login with DID + behavioral proof
The SEV aggregates features across platform categories:
Social Networks
- Account age (bucketed)
- Ego graph density
- Posting regularity entropy
- Relationship churn rate
Messaging Platforms
- Distinct conversation peers
- Reciprocity metrics
- Response latency distribution
- Conversation longevity
Content/Activity Services
- Consumption diversity entropy
- Longitudinal engagement
- Device stability
- Routine consistency
- RP sends challenge nonce + SEL policy requirements
- User's SEL agent computes current SEV and continuity score
- Agent generates ZK proof that SEV satisfies policy
- Agent signs challenge with DID private key
- RP verifies signature + ZK proof without learning raw SEV
A DID is a globally unique identifier (like did:key:z6Mk...) that resolves to a document containing public keys and service endpoints, without requiring a central registration authority.
SEL-DID integrates behavioral evidence directly into the DID structure as user-controlled credentials, rather than treating them as external attestations. No single operator controls the system.
No. SEL-DIDs explicitly maintain a high-entropy cryptographic root. Social entropy is a layered signal for Sybil resistance and recovery, not key derivation material.
SEL-DIDs are designed for platform failures. No single platform is mandatory, and the system continues functioning with reduced (but still useful) behavioral signals.
The design minimizes this risk through feature reduction, quantization, zero-knowledge proofs, and user-controlled storage. See the paper's ethics section for detailed analysis.
New users naturally have low social entropy. SEL-DIDs are optional layers; bare DIDs work in contexts not requiring Sybil resistance, and alternative evidence (community attestations, offline ceremonies) can supplement thin histories.
@article{barak2024seldid,
title={Social Entropy Layered DIDs (SEL-DIDs): Strengthening Decentralized Identity with Cross-Platform Behavioral Proofs},
author={Barak, Sahar},
year={2024}
}- W3C DID Core Specification
- Verifiable Credentials Data Model
- Decentralized Web Nodes (DWN)
- Web5 Architecture
- Proof of Personhood
- Sybil Attack
decentralized identity, DID, self-sovereign identity, SSI, sybil resistance, proof of personhood, verifiable credentials, zero-knowledge proofs, Web5, decentralized web nodes, DWN, passwordless authentication, account recovery, social entropy, behavioral biometrics, cross-platform identity, privacy-preserving identity, BrightID alternative, Gitcoin Passport alternative, Worldcoin alternative, decentralized authentication, Web3 identity
Sahar Barak Independent Researcher hi@saharbarak.dev
This work is provided for academic and research purposes.