Neural Vision
Neural Vision

Solving the Document-Based DLP Problem

Data Loss Prevention (“DLP”) objectives commonly include pinpointing documents which are restricted, confidential and toxic as defined by security policies, in order to protect an organization from material harm resulting from unauthorized possession and use.

These documents can be sitting “dark” on file servers, email servers and other content management repositories. This means they may have no metadata describing what they are, and are often images with no searchable text.

Until now, only text-based search and analytics methods were available, which are prone to precision and recall weaknesses and poor recognition results.

Current State Using
Text-Based Methods

Current State Using Text-Based Methods
  • Risk in attachments
  • Problematic file extensions
  • Detection gaps

Using NeuralVision technology, these problems are overcome because our visual signature is impervious to whether or not there is text available to search and analyze. This means the riskiest documents can be seen and protected.

Proposed State using
NeuralVision Technology

Current State Using Text-Based Methods
  • Less risk in attachments
  • Increased file type coverage
  • Enhanced detection

Visual Signatures – Defined

Visual signatures are analogous to a human brain remembering what a document looks like, and using that intelligence to look for other documents which look the same. These signatures are algorithmically generated and are based on known examples of target data provided by the client, as well as discovered versions uncovered during processing.

In the self-organizing mode, we compare visual signatures on the fly and produces clusters of same-looking documents which can be tagged with security level and other functional metadata.

Visual Signatures - Defined

Ways to Use

There are two ways to use NeuralVision technology.

1. Feed NeuralVision examples of electronic documents (file types including TIF, PDF, MSOffice and JPEG) which are confirmed to be on the target list for DLP. This would include such content as contracts, financials, design documents, intellectual property and other types of data which the organization seeks to protect. Our software generates visual signature models representing those documents. Execute the “seek” command to locate other documents which match the visual signature model.

2. In addition to or as an option to the above, let the NeuralVision software self-organize the entire population of untagged documents on the fly. Users can browse these clusters of same document types and select and tag those documents as appropriate. Tagged documents can be added to the training sets and used for future scans as necessary.

Before

Before

After

After

NeuralVision technology works equally well on small and large document collections, and persists tags to future data sets.

Deployment Options

NeuralVision technology can be used either as a managed cloud service, leveraging our SOC2-certified cloud computing environment, or as an on-prem command line-driven sub-process. Visual models are binary entities and are portable, within or across projects. For email systems, NeuralVision can process the attachments to messages and determine if that attachment matched the visual signature of a known security threat.

Cloud Hosted
Feature Benefit
Intuitive and Automatic Text-based methods rely on highly complex algorithms that are difficult for people to interpret and hard to describe to others. Our solution automatically places all documents of the same type into a folder by comparing visual signatures across current and prior collections, it uses visual recognition the same way people do, which means it’s easy to understand.
Spend Less, Do More Text-based methods require large amounts of human capital for training and QC to achieve 99+% accuracy, which is cost prohibitive. Images very hard to process. Works well with images.
Predictable Consistency Text-based methods often produce different results when the same document goes through the process multiple times. We generate an identical signature when the same document goes through the process multiple times. A high level of repeatable precision means reliable confidence.

About Us

NeuralVision Technologies LLC is a software company based in Boston, MA whose mission is to develop novel solutions in the computer and machine vision space to solve document recognition and classification problems. For more information, please visit our website www.neuralvision.net or email info@neuralvision.net.