Verification Methodology
The principles, cryptographic provenance checks, and editorial standards governing every record published on ClaimKhoj India.
1. Deterministic Source Ingestion
We ingest public notifications using deterministic crawler workers that target configured public source families. Each crawled document is hashed (SHA-256) and paired with its canonical HTTP source URL and server timestamp. This ensures every extracted datum has an auditable origin trail.
2. Structured Parameter Extraction
Legal documents are parsed into structured database candidate entities consisting of:
- Entity Identification: Exact legal corporate name, CIN, or regulatory registration identifier.
- Affected Scope: Clearly bounded group of consumers, investors, or creditors.
- Submission Deadline: Explicit cutoff date parsed in Indian Standard Time (IST).
- Action Portal: The direct HTTPS endpoint hosted by the official regulator or statutory administrator.
3. The Human Editorial Gate
Automated extraction alone is never trusted for publication. A trained editorial reviewer manually compares each candidate record against the official order text. A record is only marked is_published = true when:
- The source URL resolves to an authentic government or regulatory domain.
- The claim action route is active and does not charge unofficial intermediary fees.
- The eligibility summary accurately reflects the official criteria without hyperbole.
4. Freshness and Lifecycle Tracking
Records in ClaimKhoj undergo regular re-verification sweeps. If an official deadline passes, the record status transitions immediately to closed or expired. If a deadline is officially extended by a court, the record is updated with an editorial change note.