– virtual intelligence architecture

It takes an abundance of confidence to decide to build a suspension bridge that’s fifty percent longer than the longest one in existence. When completed in 1883, the Brooklyn Bridge was the longest suspension bridge in the world, and would remain so for 20 years.

virtualworks has taken on a lofty challenge with its virtual intelligence architecture (via): extract real value from enterprise data by understanding human-generated content (the 80% of enterprise data that is free form and ignored by other BI companies).

via’s goal is to provide the platform that enables actionable business insights can be surfaced, in real-time. Thus, embedding intelligence into workflow that helps organizations to make informed decisions, streamline processes and drive revenue.

via is the foundation of all products at virtualworks. It combines best of breed open source search and data technology with our proprietary super-performant language analysis and information extraction frameworks.

This platform enables powerful information discovery applications like viasuggest, viaworks, vialocate and viainsight. These products help users navigate to and find content and help businesses leverage their data for insight.

With the via Lexicon our clients get access to high-coverage electronic dictionaries that help recognize all the different forms in which a word may appear. via Lexicon is available for all major European languages.

Verticals or domains such as fashion, travel, consumer electronics or medicine require specialized dictionaries for proper query analysis and in-depth text mining. Through collaboration with industry experts, virtualworks has gathered many man years' worth of specialized semantic resources. These resources include keyword lists, synonym and hyponymy tables and disambiguation rules.

  • Extensive out-of-the-box dictionaries for many verticals and domains
  • Proven and manageable processes to update and enrich these dictionaries
  • Processes to detect and harness reference lists available on the Web


The only way to correctly and efficiently extract the right information

Integrate information from every corner of your digital landscape. Connect to, index and search your CMS, DMS, CRM, email, business application, and so much more. Don’t see your system? We build custom connectors.

More details about our connectors:

Out-of-the-box Databases


The only way to correctly and efficiently extract the right information

Out of the box, via integrates with Microsoft Active Directory.


Unlock the hidden data in scanned documents

With powerful optical character recognition (OCR), via fully indexes previously non-searchable PDF, TIFF, JPG, FAX and GIF files.

"Local grammars"

The only way to correctly and efficiently extract the right information

In contrast to popular approaches to syntax and parsing, virtualworks is strongly committed to the use of extremely large local grammar systems which can analyze substantial portions of natural language with a very high degree of accuracy.

Since we are still a far cry from a comprehensive and semantically realistic treatment of large fragments of any natural language, virtualworks aims rather at describing well-understood semantic subsets in a detailed way. This holds for the description of both argument and predicates. The latter express relations between arguments (virtualworks calls them “propositional forms”) and they come in a number of very different forms that are however rarely distinguished in other approaches to syntax.

Local grammars for propositional forms have the advantage that they identify the nature of the predicates by separate grammars for the different ways of expressing propositional forms (thereby also capturing the many variants of the same underlying propositional form). On the other hand, there are also very detailed grammars for the different types of argument structures involving such notions as persons, dates, organizations, locations and many others, each which requires a grammar specification of its own.

General rule based grammars that do not make such semantic distinctions from the beginning cannot capture the structure of utterances in a reliable way.