Ghost Data Detecor

Inspiration

Across industries like government, healthcare, logistics, and insurance — corruption or inefficiency often doesn’t come from wrong data, but missing data. Records disappear intentionally, forms are never created, or digital entries are “lost” with no trace. These invisible gaps hide fraud, slow delivery of public services, and weaken accountability. We wanted to shine a light on the data that should exist but doesn’t — the hidden truth.

What it does

Ghost Data Detector scans datasets to identify:

Missing records that should logically exist
Broken chains of documentation (e.g., issue raised → no resolution logged)
Suspicious gaps in time-based or sequential data
Unlinked entities where mandatory fields are empty or removed It detects patterns of ghost data using relational rules, anomaly tagging, and audit trail reconstruction. It helps auditors, journalists, and organizations spot fraud or inefficiency quickly.

How we built it

We built an intelligent rule-based and AI-assisted system that:

Maps entity relationships and dependency chains in data
Uses logic checks to identify missing data ("If A exists, B must exist")
Applies anomaly flags when sudden blank zones appear
Generates reports highlighting suspected tampering areas Designed as a plug-and-play module for datasets across multiple sectors.

Challenges we ran into

Hard to prove intentional data deletion vs. natural absence
Different departments store data in different formats
High sensitivity around exposing corruption
Need for scalable logic that works across industries

Accomplishments that we're proud of

Built a detection system for something normally ignored: missing data
Identified patterns that can uncover fraud and negligence
Designed a tool that supports public transparency and justice
Created a framework that scales across domains — governance, hospitals, insurance, logistics, and more