This outline is provisional and subject to change.
SAS Visual Text Analytics enables you to uncover insights hidden within unstructured data using the combined power of natural language processing, machine learning, and linguistic rules. This course explores the five components of Visual Text Analytics: parsing, concept derivation, topic derivation, text categorization, and sentiment analysis. Documents are parsed and analyzed to reveal dominant themes in the document collection. Sophisticated linguistic queries are constructed to satisfy specific information needs. An integrated solution is developed using information extracted from subject matter expert rules, combined with machine learning results for model and rule-based topics and categories. The course includes hands-on use of SAS Viya in a distributed computing environment.
Learn how to
- Use the point-and-click interface of Model Studio and SAS Visual Text Analytics.
- Explore collections of text documents to discover key topics.
- Interpret term maps.
- Identify key textual topics automatically in your large document collections.
- Create robust models for categorizing the content according to your organization's specific needs.
- Create, modify, and enable (or disable) custom concepts and test linguistic rule definitions with validation checks within the same interactive GUI.
- Extract individual instances of concepts from within documents.
- Create custom Boolean rules to categorize documents with respect to a categorical target variable.
- Modify automatically generated Boolean category rules.
- Extract a document-level sentiment score.
- Create modeling-ready data for use by SAS Visual Data Mining and Machine Learning.
Who should attend
Text analysts, business and marketing analysts, web analysts, BI professionals, customer intelligence professionals, social media analysts, and document librarians
This outline is provisional and subject to change.
Neither SAS programming experience nor statistical knowledge is required. You should be comfortable using a computer, have experience using browser-based software solutions, and have a basic understanding of the differences between structured (numeric) and unstructured (text) data fields.
This course addresses SAS Visual Text Analytics software.
This outline is provisional and subject to change.
Introduction to SAS Visual Text Analytics- Introduction.
- Language challenges (self-study).
SAS Visual Text Analytics Demonstrations- Importing document collections.
- Creating a project with no predefined concepts.
- A project with custom concepts.
SAS Visual Text Analytics Nodes- Introduction.
- Concepts and terms.
- Machine-generated topics.
- Categories.
- Scoring new documents.
Concept and Category Rule Definitions- SAS Visual Text Analytics rules.
- SAS Visual Text Analytics concept rules.
- SAS Visual Text Analytics demo category rules.
Case Studies- Retrieving information and documents about anxiety and depression from drug reports.
- Automatic categorization of ASRS incident reports.
- Retrieving mortgage complaints from the CFPB customer complaints data (self-study).
