E-Discovery Services
  » Collection
  » Analysis
  » Processing
  » Review
  » Production

Invariant™ Overview

Invariant™ Overview:

Quick Summary

• Built from the Ground Up to Support the Unicode Standard
• Foreign Language Extraction and Precision OCR (122 languages)
• Support for Deeply Embedded Documents and Images
• Open Software Architecture Allows Rapid Customization
• Scalable System – Relieves Bottlenecks With Massive Throughput
• Dynamic Searching and Filtering With No Reloads Required
• Recognizes Over 2,500 File Types From Multiple Operating Systems
• Extremely Accurate Rendering Independent of Print Driver Limitations
• Export as TIFF or Searchable PDF (Full Color with Interactive Links)

Technical Overview

System Features

• Fully Distributed Processing
  Workloads for projects are shared across multiple multi-threaded worker machines. Discovery, data extract, TIFF'ing and OCR are all distributed and load balanced.
  
  
• Full Unicode Support
  Designed for Unicode from the ground up.  Original client data which is not in Unicode is up-converted to Unicode.  Data delivery to 3rd party applications can be accomplished in Unicode, ANSI, RTF or more than two dozen other text encodings depending upon the application requirements of the end user.
  
  
• Internationalized Full-Text Search
  Full text searches on extracted text can currently be performed in 122 languages.

Discovery

• File Identification
  Invariant can identify over 2,500 types of files from multiple operating systems and examines file content rather than file extension to determine file type.
  
  
• Metadata
  Invariant never assumes a predefined set of metadata fields. Instead, our software first walks files in order to preserve volatile metadata while examining and capturing any/all metadata found within native files. Once new metadata fields are discovered they are given a unique ID for cataloging and added to our progressive metadata library (currently more than 12,000 metadata fields).
  
  
• File Formats
  Invariant™ currently supports over 2,500 file types including all major email and data files. Below are just a few examples:
  
  • All standard files and file types
  • Adobe Acrobat including packages and portfolios
  • Bloomberg emails
  • Microsoft Snapshot files
  • Can extract CAB file archives
  • Can extract RAR compressed files
  • Full Office 2007 support including customized extraction and rendering
  • XPS fixed and flowed documents
  • Can extract annotations, comments and attachments from PDF files
  • Can typically address rare or exotic file types with custom code in 2 days
  
  
• Deep Embedding
  Invariant uncovers deeply embedded objects and sub-documents (for example, an email with attachments dragged and dropped into a Word document; an embedded Word document inside an Excel spreadsheet) through a process of infinite recursion.
  
  
• Dates & Times
  Dates and times are stored in UTC. Time zone adjustments can be made during the data export phase rather than having to preset the time zone before discovery begins.
  
  
• TIFF'ing and OCR
  The OCR engine is only utilized on pages where an image is present. If a page of a document contains a mixture of text and graphics, the text is extracted separately and then the graphics are OCR'd. The resulting text for that page will physically separate what was OCR'd versus what was text extracted.
  
  OCR engine supports 122 languages, including Chinese Traditional, Chinese Simplified, Korean and Japanese.

Exporting & Load Files

• Highly Customizable
  Invariant’s powerful export features allow the combination of export tasks or split export tasks into separate work units. Load files can be built rapidly without having to copy the native files or images, allowing for quick work verification. This technology also allows us the ability to resume an export without overwriting the existing destination files.
  
  
• Rich-Text Supported
  Invariant™ supports rich-text so complete hit-highlighting from search requests can be included in the exported text.
  
  
• Image Endorsements on TIFF/PDF
  Export documents as TIFF images or PDFs with full color and interactive links. Endorsements on PDF exports are fully-searchable and can be completely customized.
  
  
• Internationalized Output
  Export text in RTF, UTF-8, UTF-16, Unicode, ANSI or more than two dozen other text encodings.
  
  
• Wide Range of Built-in Export Definitions
  Extensible Export Mechanism means we can provide a wide range of built-in export definitions with Invariant. We deliver a customized export solution to fit any third-party application requirements including, but not limited to: Concordance®, CT Summation and Ringtail®.