In-Memory Data Pseudonymization for Research
Your dataset is never written to storage.
Processing takes place entirely in volatile memory, across every supported format.
Supported formats: CSV / TSV, Excel, XML, JSON, DICOM P-RDSR, HL7 v2.x / FHIR / SS-MIX, SWIFT, ISO 20022, UN/EDIFACT.
Each session ends with a Processing Completion & Non-Retention Certificate.
1
Upload & Parse
Upload your structured data through the browser interface. Edge scanners check the file structure locally before upload.
2
In-Memory Pseudonymization
The engine identifies designated tags and replaces sensitive elements with consistent pseudonyms, operating entirely in volatile memory.
3
Download & Purge
Retrieve your pseudonymized dataset and a system-generated Non-Retention Certificate. The in-memory state is purged when the session ends.
Core Workflow
A Format-Agnostic, Database-Free Architecture
A processing pipeline that runs in memory and is independent of file format. Unlike conventional tools that write temporary files to server storage, Yata avoids server-side persistence by design: the in-memory state is purged when the session ends, and a Non-Retention Certificate is issued on completion.
Any supported input
In-Memory Pseudonymization
Pseudonymize → Export
Memory Purge
Structural Inspection & Pseudonymization
PII Pseudonymization
Identify and pseudonymize sensitive columns — such as sender names and transaction IDs — with a toggle. All operations run in memory.
Data Inspection
Inspect, filter, and analyze structured messages such as SWIFT and ISO 20022 in a table view.
Risk & Anomaly Simulation
Calculate heuristic risk scores and detect statistical deviations using methods such as Benford's Law analysis. Intended for research and testing.
Visual Analytics
Explore transaction relationships with interactive network maps, or identify data clusters with PCA-based scatter plots (Galaxy View).
Secure Export & Session Finalization
Select Finalize Session & Export to download your pseudonymized dataset together with a system-generated Non-Retention Certificate. The in-memory state is purged when the session ends.
Engineered for Secure Data Workflows
Medical Informatics
DICOM P-RDSR / HL7 v2.x / FHIR
Tag-level pseudonymization of imaging and clinical records. Supports researchers in preparing pseudonymized datasets for academic and testing use without writing to persistent storage.
Financial Infrastructure
SWIFT / ISO 20022
Pseudonymization of PII in SWIFT and ISO 20022 messages. Values are replaced consistently across files, and the Key Table records the mapping so that results can be traced within your own environment. Because the mapping exists, the output is linkable pseudonymized data, not anonymized data.
Enterprise & Logistics
CSV / TSV / Excel / XML / JSON / UN/EDIFACT
In-memory pseudonymization of CSV, TSV, Excel, XML, and JSON records. Suited to generating pseudonymized test data for supply chain and logistics development.