Natural Language Processing

Autoclassification enhanced with Natural Language Processing

BA Insight's AutoClassifier has been integrated with best of breed AI capabilities from Microsoft and Google, bringing advanced capabilities that help users find information faster and surface automated intelligence about all content.

AutoClassifer is available for Elasticsearch, Azure Search, DataFrameworks ClarityNow, Box, and SharePoint Online/On-premise.


The key capabilities of AutoClassifier's Natural Language Processing features are:

Automatic creation of summarizations of key documents

  • End users are provided with a summary of specifically what the document is about, allowing them to rapidly determine if the document meets their needs
  • Relevancy is tuned to recognize hits against this summary and returns these matches with greater weight than full text hits, greatly increasing the quality of the initial search results

Identification of similar documents

  • Utilizing extraction and comparison of key concepts from each document, users are automatically presented similar documents, even when those documents do not contain the exact phrase that was searched

Extraction of concepts found within the document

  • Automatically identify locations, people, and organizations referenced within the document and leverage this as metadata to drive search
  • Easily extend these concepts to include organization specific names, departments, products, and more

Extends to all enterprise data

  • Based on out of the box integration with BA Insight's ConnectivityHub, this capability applies to any connected data source

Benefits of Natural Language Processing

AutoClassifier's Natural Language Processing is based on best of breed machine learning, which delivers many benefits:

Users are more likely

to find what they are looking for as simple keyword searching is replaced with intent-based searching

Machine Learning

automatically focuses on common cases whereas when writing rules manually it is often not obvious where the efforts should be directed

Machine Learning

can make use of statistical algorithms to produce models that are robust to unfamiliar input (e.g. containing words or structures that have not been seen before) and to erroneous input (e.g. with misspelled words or words accidentally omitted)

The Natural Language Processing

can be made more accurate simply by supplying more input data