We’re smarter together. Learn from this collection of community knowledge and add your expertise.

SAS Text Analytics Tools: An Overview

by Contributor JuliaM on ‎02-06-2013 07:16 PM - edited on ‎10-05-2015 03:02 PM by Community Manager (578 Views)

In response to a question that we received from a participant, I have written up a brief overview of the tools that are mentioned in our online poll on SAS tools. The header links go to the appropriate places on the SAS website, in case anyone wants to read more about the tools. I am most familiar with the Enterprise Content Categorization tool, so if anyone has experience in using the other tools, please share your thoughts about how you use the other tools. 


SAS Enterprise Content Categorization uses a combination of both linguistic rules and statistical analysis to classify documents using either hierarchical or flat taxonomies and to extract key information. There are two main components to the SAS Content Categorization.


  • Categorization

The Categorization component will classify documents using the criteria that the user chooses. Users can classify documents using algorithms to determine patterns, manually create Boolean rules based on pre-existing taxonomies or a combination of both.

  • Concepts extraction

The Concepts extraction will allow users to extract information that might not be known beforehand. For example, the Concepts extraction component can be set to extract the names of companies from a set of documents based on pre-set criteria.




SAS Ontology Management is used to create ontologies or taxonomies of terms  that are applied by SAS Categorization to classify documents - as well as the definitions of the concepts that can be extracted. One of the useful functions of the SAS Ontology Management is the ability to allow different users to create different taxonomies to suit their needs, but to store the terms in a central repository. This allows users from different parts of an organization to see what terms the other is using in their sections.



SAS Text Miner discovers patterns in text collections which may not be readily apparent, and can generate a quantitative representation of text, which can be used in conjunction with data mining algorithms, adding text- variables, to traditional structured data analysis. A case study of the Louisville Hospital shows how this tool can be used in the medical field.



SAS Sentiment Analysis searches for and evaluates internal and external content about a user’s organization or their competitors by identifying positive, negative, neutral and "no sentiment" details within texts and then quantifying these texts to gauge perceptions about the organization. Using a combination of statistical modeling and rule-based natural language processing the Sentiment Analysis software can extract sentiments in real time or over a period of time to show patterns and detailed reactions.

Your turn
Sign In!

Want to write an article? Sign in with your profile.

Looking for the Ask the Expert series? Find it in its new home: communities.sas.com/askexpert.