Combine machine learning with subject-matter expertise to identify trends and topics within unstructured data. You guide the technology to define categories that are important to your organization, and bring order to the chaos of free-format text.
Classify documents for better retrieval and reporting.
Easily structure documents, emails, books and other text sources to find key categories within text information. Then classify documents and format scored data for use directly in SAS Visual Analytics or other SAS applications.
Eliminate lengthy, error-prone manual reviews.
Remove the inconsistency of manual tagging so you can process higher volumes of information with better quality by applying derived text models to documents. And reduce the costs of using valuable subject-matter experts for standard tasks.
Automate taxonomy discovery for categorization.
Instead of manually defining a training corpus to initiate model development, SAS Contextual Analysis uses machine-learning methods to define themes in unstructured data – for a faster, data-driven start to taxonomy development.
Refine rules to enhance text analysis accuracy.
A single interface allows you to review categories identified by machine learning. Then you can edit, augment and clarify discovered rule models and validate changes by testing them against validation samples.
- Integrated system that guides categorization model development.
- Hybrid approach to classifying documents.
- Direct integration with SAS.
- Natural language processing (NLP).
- Automatic discovery of topics.
- Configurable categorization rule generation.