Text Analytics

Knowledge Management taken to the next level using advanced data science. Using text analysis algorithms including named entity extraction and ontological classification, your organization can leverage the digital knowledge created by your employees over the last decades to ensure it stays productive and cutting edge. 

How it can help your organization

  • Mitigate risks of retiring workforce and employee turnover
  • Leverage analytics to grow operational capability
  • Ensure leadership and employees alike have access to the most current and relevant information
  • Decrease learning curves and improve employee transition

How it works

Extraction, Translation, and Loading make your existing archives, emails, reports, and other digital media to accessible and searchable.

Natural Language Processing algorithms append metadata that provide additional search and pre-processing to speed up search capability.

Foreign language processing ensures your entire organization, regardless of nationality, leverages the same digital knowledge

Named Entity Extraction uses metadata to identify actor(s), regions(s), function(s), etc. beyond search engine like indexing

Link Analysis connects documents, texts, and subjects that provide real digital knowledge to employees and users. 

Ontological Classification algorithms classify your digital knowledge in the way that makes the most sense for your organization such as actor(s), region(s), and function(s). 

Application Programming Interfaces (APIs), Resource Description Framework (RDF) databases, and desktop tools load, process, extract, link, and classify streaming data to keep your organization cutting edge

History and Challenges

 Digital Knowledge Analytics algorithms were originally designed to build a next-generation intelligence collection and analysis grid to support ongoing intelligence services.  In building these algorithms it quickly became apparent that extracting meaningful intelligence from open source and unstructured collections presented several challenges including unique lexicon(s) and set(s) of actors and frequent interleaving of human readable text and computer-represented data, such as source code.  Working through these and other challenges one by one it quickly became apparent how useful these tools could become for any organization. Analyze developed a simple approach to tailor these algorithms to businesses and government organizations archives and historic documents to help address a growing need to mitigate knowledge and experience loss that occurs from a retiring workforce and employee turnover. 


Speak with a data scientist