SAS® Visual Text Analytics

Uncover insights hidden within unstructured data using the combined power of natural language processing, machine learning and linguistic rules. 

SAS Visual Text Analytics shown on three desktop monitors

Get maximum value from your unstructured data using a wide variety of modeling approaches – including supervised and unsupervised machine learning, linguistic rules, categorization, entity extraction, sentiment analysis and topic detection. SAS Visual Text Analytics helps you overcome the challenges of identifying and categorizing large volumes of text data. 

SAS Visual Text Analytics showing text sentiment analysis on desktop monitor

Detect emerging trends and hidden opportunities. 

Automatically convert unstructured data into meaningful insights that feed machine learning and predictive models, so you can surface new opportunities – and act on them – quickly. The software combines machine-learning methods with a rules-based approach that's essential for understanding the subtle nuances of language and inferring intention. You can also add subject-matter expertise to further increase the accuracy of your text models.

SAS Visual Text Analytics showing topic matches on desktop monitor

Drive faster results with automation. 

Intelligent algorithms and NLP techniques automatically identify relationships and patterns in text data, eliminating time-consuming manual analysis. Identifying and extracting important topics in free-form text produces new variables that enhance predictive models, reports, and search or filtering applications.

SAS® Visual Text Anaytics - text topics output in SAS® Visual Analytics decision tree

Go from data to decisions faster. 

Empower decision making at the source of the data, and reduce the gap between when information is received and when it's acted on. If someone leaves a comment or clicks through an app on a mobile device, SAS Visual Text Analytics analyzes the data immediately using in-memory, in-database and in-stream technologies. Analyzing real-time streaming text data dramatically speeds up the data-to-decision timeline. 

SAS Visual Text Analytics default pipeline shown on desktop monitor

Foster collaboration and information sharing in an open ecosystem. 

SAS Visual Text Analytics provides a flexible environment that supports the entire analytical life cycle – from data preparation and visual exploration to analysis and deployment. You can tackle and experiment with a variety of analytical use cases to support a single initiative. Collaboration is possible at all levels, whether you’re a data scientist preparing data, a domain expert applying linguistic rules or an IT person deploying models. This unified solution also integrates seamlessly with existing systems and open source technologies.

Key Features

  • Combined machine-learning and rules-based methods. Automatically identify core themes in a collection of documents using machine-learning techniques, and apply linguistic rules to understand slang and infer intention. Create precise, rules-based categories and concepts when discovered themes need additional tuning or custom definitions.  
  • Contextual extraction. Detect and extract data elements and relationships from unstructured text with a comprehensive toolset that includes predefined concepts, as well as the ability to create custom concepts and definitions. 
  • Flexible deployment. Deploy models in batch, in Hadoop, in stream and via APIs. Run models closer to where data is collected to reduce data movement and produce faster results for scoring new data. 
  • Multiuser environment. The SAS Platform offers more depth and breadth of integrated analytical capabilities than any other software. This integration encourages teamwork and collaboration by providing a workspace for sharing best-practice pipelines and methods. 
  • Natural language processing. Automatically analyze and transform text into formal representations for text processing and understanding using NLP. Natural language understanding (NLU), a subset of NLP, enables contextual understanding of content.
  • Automated machine-generated topic detection. Derive topics automatically from your documents using two unsupervised machine-learning methods – singular value decomposition and latent Dirichlet allocation. 
  • Native support for 30 languages. Eliminate the need to translate language prior to analysis with proprietary language packs that enable native language analysis using dictionaries and linguistic assets created by native language experts. 
  • Sentiment analysis. Identify and analyze terms, phrases and character strings that imply an author’s tone or attitude (positive, negative or neutral) expressed through text.
  • Open APIs. Seamlessly integrate with existing systems and open source technologies, and add the power of SAS Analytics to other applications using SAS® Viya® REST APIs.
SAS Viya

 

To rozwiązanie jest oparte na SAS Viya - nowoczesnej platformie, zdolnej sprostać każdemu wyzwaniu analitycznemu. SAS Viya, jako zunifikowane i gotowe do działania w chmurze środowisko, obsługuje wszystkich - od specjalistów data science i analityków biznesowych po deweloperów aplikacji i kadrę zarządzającą - dzięki niezawodnemu, skalowalnemu, bezpiecznemu zarządzaniu analityką oraz nadzorowi niezbędnemu w sprawnie funkcjonującym środowisku IT. Oferuje ono także wydajność, jakiej można oczekiwać od SAS - światowego lidera w dziedzinie analityki.

Recommended Resources


Get more details in the SAS Visual Text Analytics fact sheet.


Discover what text analytics can do for your organization in this research brief from the International Institute for Analytics.

Get white paper


Learn how any size organization can supersize results with data visualization.

Download e-book

Back to Top