InfoWatch has released version 2.2 of its Data Discovery solution, featuring a new machine learning module for automated file clustering. This tool analyzes 100% of documents in an IT infrastructure without requiring manual dictionary setup or pre-defined policies.

The system automatically categorizes files by topic and assigns tags, which the company claims can accelerate auditing processes by up to 30 times compared to traditional manual methods. Indexing occurs in the background, allowing for instant search and filtering across the entire storage environment.

The new functionality helps eliminate 'gray zones,' which previously accounted for up to 70% of uncontrolled data. Security specialists can use the clustering results to identify rare or unauthorized documents and track the emergence of new data categories over time.

The module integrates with the InfoWatch Traffic Monitor DLP system. Users can convert document clusters into dictionaries with a single click, reducing the configuration time for new data categories from 3–5 days to just a few minutes.