SAS® Unified Insights MM Features
Market-leading data mining & machine learning
- Provides GUI-based data mining and machine learning via a single, collaborative and highly scalable environment.
- Provides open source integration with R, Python, Java and Lua models.
- Lets you use model competition to identify and deploy the most effective model.
View more market-leading data mining & machine learning features
Streamlined model deployment
- Streamlines the process of creating, managing, administering, deploying and monitoring your analytical models.
- Provides a framework for model registration, validation, monitoring and retraining.
- Enables you to assess candidate models to identify and publish the champion model.
- Ensures complete auditability and regulatory compliance.
View more streamlined model deployment features
- Provides secure, reliable, versioned storage for all types of models, as well as access administration, including backup and restore capabilities, overwrite protection and event logging.
- Once registered, models can be searched, queried, sorted and filtered by attributes used to store them – type of asset, algorithm, input or target variables, model ID, etc – as well as user-defined propertied and editable keywords.
- Add general properties as columns to the listing for models and projects, such as model name, role, type of algorithm, date modified, modified by, repository location, description, version and keywords (tags).
- Access models and model-score artifacts using open REST APIs.
- Directly supports Python models for scoring and publishing. Convert PMML and ONNX (using dlPy) to standard SAS model types. Manage and version R code like other types of code.
- Provides accounting and auditability, including event logging of major actions – e.g., model creation, project creation and publishing.
- Export models as .ZIP format, including all model file contents for movement across environments.
- Easily copy models from one project to another, simplifying model movement within the repository.
Analytical workflow management
- Create custom processes for each model using SAS Workflow Studio:
- The workflow manager is fully integrated with SAS Model Manager so you can manage workflows and track workflow tasks within the same user interface.
- Import, update and export generic models at the folder level – and duplicate or move to another folder.
- Facilitates collaboration across teams with automated notifications.
- Perform common model management tasks, such as importing, viewing and attaching supporting documentation; setting a project champion model and flagging challenger models; publishing models for scoring purposes; and viewing dashboard reports.
- Place a combination of Python, SAS or other open source models in the same project for users to compare and assess using different model fit statistics.
- Set up, maintain and manage separate versions for models:
- The champion model is automatically defined as a new version when the model is set as champion, updated or published in a project.
- Choose challenger models to the project champion model.
- Monitor and publish challenger and champion models.
- Define test and production score jobs for SAS and Python models using required inputs and outputs.
- Create and execute scoring tasks, and specify where to save the output and job history.
- Depending on the use case, you can publish models to batch/operational systems – e.g., SAS server, in-database, in-Hadoop/Spark, SAS Cloud Analytic Services (CAS) Server, or to on-demand systems using Micro Analytic Score (MAS) service.
- Publish Python and SAS models to run time containers with embedded binaries and score code files. Promote run time containers to local Docker, AWS Docker and Amazon EKS (elastic kubernetes service) environments.
- Monitor the performance of models with any type of score code. Performance reports produced for champion and challenger R, Python and SAS models include variable distribution plots, lift charts, stability charts, ROC, K-S and Gini reports with SAS Visual Analytics using performance-reporting output result sets.
- Built-in reports display the measures for input and output data and fit statistics for classification and regression models to evaluate whether to retrain, retire or create new models. Performance reports for champion and challenger analytical models involving Python, SAS, R, etc., with different accuracy statistics are available.
- Monitor performance of champion models for all projects using performance report definition and execution.
- Schedule recurring and future jobs for performance monitoring.
- Specify multiple data sources and time-collection periods when defining performance-monitoring tasks.
Self-service data preparation
- Provides an interactive, self-service environment for data access, blending, shaping and cleansing to prepare data for analytics and reporting.
- Fully integrates with your analytics pipeline.
- Includes data lineage and automation.
View more self-service data preparation features
Data & metadata access
- Use any authorized internal source, accessible external data sources and data held in-memory in SAS Viya.
- View a sample of a table or file loaded in the in-memory engine of SAS Viya, or from data sources registered with SAS/ACCESS, to visualize the data you want to work with.
- Quickly create connections to and between external data sources.
- Access physical metadata information like column names, data types, encoding, column count and row count to gain further insight into the data.
- View a sample of a table or file loaded in the in-memory engine of SAS Viya, or from data sources registered with SAS/ACCESS, to visualize the data you want to work with.
- Data sources and types include:
- Amazon S3.
- Amazon Redshift.
- DNFS, HDFS, PATH-based files (CSV, SAS, Excel, delimited).
- SAS® LASR™.
- Feeds from Twitter, YouTube, Facebook, Google Analytics, Google Drive, Esri and local files.
- SAS® Cloud Analytic Services (CAS).
- Parallel load data from desired data sources into memory simply by selecting them – no need to write code or have experience with an ETL tool. (Data cannot be sent back to the following data sources: Twitter, YouTube, Facebook, Google Analytics, Esri; it can only be sourced form these sites).
- Reduce the amount of data being copied by performing row filtering or column filtering before the data is provisioned.
- Retain big data in situ, and push processing to the source system by including SAS In-Database optional add-ons.
Guided, interactive data preparation
- Transform, blend, shape, cleanse and standardize data in an interactive, visual environment that guides you through data preparation processes.
- Easily understand how a transformation affected results, getting visual feedback in near-real-time through the distributed, in-memory processing of SAS Viya.
Machine learning & AI suggestions
- Take advantage of AI and machine learning to scan data and make intelligent transformation suggestions.
- Accept suggestions and complete transformations at the click of a button. No advanced or complex coding required.
- Automated suggestions include:
- Gender analysis.
- Match code.
- Missing value imputation for numeric variables.
- One hot encoding.
- Remove column.
- Whitespace trimming.
- Convert column data type.
- Center and scale.
- Unique ID creation.
- Column removal for sparse data.
- Use column-based transformations to standardize, remediate and shape data without doing configurations. You can:
- Change case.
- Convert column.
- Trim whitespace.
- Custom calculation.
- Support for wide tables allows for the saving of data plans for quick data preparation jobs.
- Use row-based transformations to filter and shape data.
- Create analytical-based tables using the transpose transformation to prepare the data for analytics and reporting tasks.
- Create simple or complex filters to remove unnecessary data.
- Write custom code to transform, shape, blend, remediate and standardize data.
- Write simple expressions to create calculated columns, write advanced code or reuse code snippets for greater transformational flexibility.
- Import custom code defined by others, sharing best practices and collaborative productivity.
- Use multiple-input-based transformations to blend and shape data.
- Blend or shape one or more sets of data together using the guided interface – there’s no requirement to know SQL or SAS. You can:
- Append data.
- Join data.
- Transpose data.
- Profile data to generate column-based and table-based basic and advanced profile metrics.
- Use the table-level profile metrics to uncover data quality issues and get further insight into the data itself.
- Drill into each column for column-level profile metrics and to see visual graphs of pattern distribution and frequency distribution results that help uncover hidden insights.
- Use a variety of data types/sources (listed previously). To profile data from Twitter, Facebook, Google Analytics or YouTube, you must first explicitly import the data into the SAS Viya in-memory environment.
Data quality processing
(SAS® Data Quality in SAS® Viya® is included in SAS Data Preparation)
- Use locale- and context-specific parsing and field extraction definitions to reshape data and uncover additional insights.
- Use the extraction transformation to identify and extract contact information (e.g., name, gender, field, pattern, identify, email and phone number) in a specified column.
- Use parsing when data in a specified column needs to be tokenized into substrings (e.g., a full name tokenized into prefix, given name, middle name and family name).
- Derive unique identifiers from match codes that link disparate data sources.
- Standardize data with locale- and context-specific definitions to transform data into a common format, like casing.
- Analyze column data using locale-specific rules to determine gender or context.
- Use identification analysis to analyze the data and determine its context, which is particularly valuable if the data or source of data is unfamiliar.
- Use gender analysis to determine the gender of a name using locale-specific rules so the data can be easily filtered or segmented.
- Create a unique ID for each row with unique ID generator.
- Identify the subject data in each column with identification analysis.
- Identify, find and sort data by tagging data with columns and tables.
- Determine matching records based upon locale- and context-specific definitions.
- Easily identify matching records using more than 25 context-specific rules such as date, address, name, email, etc.
- Use the results of the match code transformation to remove duplicates, perform a fuzzy search or a fuzzy join.
- Find like records and logically group together.
System & job monitoring
- Use integrated monitoring capabilities for system- and job-level processes.
- Gain insight into how many processes are running, how long they’re taking and who is running them.
- Easily filter through all system jobs based on job status (running, successful, failed, pending and cancelled).
- Access job error logs to help with root-cause analysis and troubleshooting. (Note: Monitoring is available using SAS Environment Manager and the job monitor application.)
Data import & data preparation job scheduling
- Create a data import job from automatically generated code to perform a data refresh using the integrated scheduler.
- Schedule data explorer imports as jobs so they will become an automatic, repeatable process.
- Specify a time, date, frequency and/or interval for the jobs.
- Explore relationships between accessible data sources, data objects and jobs.
- Use the relationship graph to visually show the relationships that exist between objects, making it easier to understand the origin of data and trace its processing.
- Create multiple views with different tabs, and save the organization of those views.
Plan templates & project collaboration
- Use data preparation plans (templates), which consist of a set of transformation rules that get applied to one or more sources of data, to improve productivity (spend less time preparing data).
- Reuse the templates by applying them to different sets of data to ensure that data is transformed consistently to adhere to enterprise data standards and policies.
- Rely on team-based collaboration through a project hub used with SAS Viya projects. The project’s activity feed shows who did what and when, and can be used to communicate with other team members.
Batch text analysis
- Quickly extract contents of documents, and perform text identification and extraction.
Cloud data exchange
- Securely copy data from on-site repositories to a cloud-based SAS Viya instance running in a private or public cloud for use in SAS Viya applications – as well as sending data back to on-site locations.
- Preprocess data locally, which reduces the amount of data that needs to be moved to remote locations.
- Use a Command Line Input (CLI) interface for administration and control.
- Securely and responsibly negotiates your on-site firewall.
Visual data exploration & insights development
- Provides bi-modal support for both governed and self-service exploration and visualization.
- Enables self-service discovery, reporting and analysis.
- Provides access to easy-to-use predictive analytics with “smart algorithms.”
- Enables report sharing via email, web browser, MS Office or mobile devices.
- Provides centralized, web-based administration, monitoring and governance of platform.
View more visual data exploration & insights development features
- Import data from a variety of sources: databases, Hadoop, Excel spreadsheets, social media, etc.
- Drag an Excel file, CSV or SAS data set onto your workspace, and quickly start building reports or dashboards.
- Use standard data quality functions like change case; convert, rename, remove and split columns; and create calculated columns and transformations using custom code.
- Prep data using append, join, filter and transpose functions.
- Reuse, schedule and monitor jobs.
- View lineage with network diagrams.
- Quickly view descriptive statistics on measures to help you see the characteristics of your data.
- Create calculated, aggregated or derived data items.
- Create drillable hierarchies in a self-service manner without the need to predefine user paths.
- Interactive data discovery enables business users and analysts to easily identify relationships, trends, outliers, etc.
- Precise and responsive layout capabilities give you flexible layout and design options. You can stack or group items, and more.
- A variety of graph objects or charts are included:
- Heat map.
- Animated bubble.
- Numeric series.
- Schedule chart.
- Key value infographics.
- And many more with flexible graph building capabilities.
- Add content from the web (e.g., YouTube videos, web apps) and images (e.g., logos) to your report.
- Custom sort allows you to rank order category data items in a table or graph by characteristics (e.g., products, customers). The characteristics that are most important to your organization will be displayed first.
- One-click filtering (e.g., one way, bidirectional) and linked selections will allow you to spend less time manually linking content (e.g., visualizations, reports).
- Text objects include date-driven or system-generated text for relevant context.
- Synchronize selection and filters across visualizations in a report or dashboard.
- Link different reports (e.g., link a sales report to an inventory report).
- Report consumers can change calculation parameters and display rules using controls, filters, etc. to see information that is most relevant to them.
- Report consumers can switch measures and change chart type and formatting all on the fly allowing them to make critical business decisions instantly.
- Set refresh rates for individual objects, pages or your entire report.
- Analytical visualizations include:
- Box plot.
- Heat map.
- Animated bubble chart.
- Network diagram.
- Correlation matrix.
- Parallel coordinates plot.
- Decision tree.
- And many more with flexible graph building capabilities.
- Geographical map views provide a quick understanding of geospatial data, including travel time and travel distance, demographics data enrichment with Esri integration.
- Network diagrams enable you to display networks across a map.
- Bring your own custom interactive visualizations (e.g., D3.js graphs, C3 visualizations or Google charts) into SAS Visual Analytics, so they’re all driven by the same data.
- Key value visualization allows you to display important metrics (numeric or categorical value) in an infographic style for quick reference.
- Perform path analysis (Sankey diagrams) to visualize relationships between a distinct sequence of events.
- Add cell visualizations, like bars and heat maps, to your tables to quickly identify problem points and see trends in your data.
- Generate forecasts on the fly with forecasting confidence intervals included.
- The most appropriate forecasting model is automatically selected after running multiple models against data.
- Scenario analysis lets you see how changes in different variables would affect forecasts.
- Goal seeking enables you to specify a target value for your forecast, and then determines the values of underlying factors that would be required to achieve the target value.
- Decision tree graphically depicts likely outcomes.
- Custom binning moves continuous data into a small number of groups for better interpretation and presentation of results.
- Text analysis capabilities enable you to automatically find topics and understand sentiment from text sources, including Facebook, Twitter, Google Analytics, YouTube comments and more.
- Recover reports you are editing when your session ends unexpectedly. Reports are automatically saved every five seconds after an edit is made.
- Pick up where you left off from a prior session on all your devices.
- Autocharting automatically chooses the graph best-suited to display selected data.
- Automated Explanation determines which variables contribute to an outcome and provides a simple natural language explanation that is easy to understand.
- Quickly detect and highlight patterns and outliers in your data with Automated Explanation.
- Automated Explanation determines the key difference between the top and bottom cases in data. For example, what best differentiates the lowest risk and the highest risk cases?
- The steps taken to automatically explain your data are displayed for transparency.
- Use Automated Explanation to identify interesting groups based on factors you select.
- Automatically builds an interactive analytical story based on all your data, ready to be published.
- Suggested insights automatically derived from your data allows you to quickly build informative reports and dashboards.
- Related measures highlighted within the measure list so users can quickly identify potential interactions.
Sharing & collaboration
- Reuse and share report modifications, such as filters, calculations, hierarchies and report element formatting.
- Collaborate across mobile devices and the web by adding comments to a report.
- Create alerts for a report object so that subscribers are notified via email or a text message when the threshold condition is met.
- Distribute reports as PDFs or email in a secure manner. Distribute reports once or at recurring intervals, such as daily, weekly or monthly.
- Playable dashboards let you put your report in slideshow mode.
- Administrators can configure support for guest access to view report or visualization.
- Guest users can view the insights that are available to the public.
- Users can see, organize and collaborate on their work using SAS Drive:
- Users can favorite, share, preview and tag their content from one place.
- Create projects that share data, content and other resources with project members.
SAS® Visual Analytics Apps
- Available for free from:
- App Store for iOS iPhone and iPads.
- Google Play for Android phones and tablets.
- Microsoft Store for Windows 10 devices.
- The app lets you connect and interact with your SAS Visual Analytics reports and dashboards using gestures native to your devices.
- Interact with your SAS Visual Analytics app for iOS using voice commands.
- Reports created once in SAS Visual Analytics can be viewed anywhere.
- Gain secure access to content on mobile devices, both online and offline.
- Annotate, comment, share and email reports to others for increased collaboration.
- Screenshots can be captured and comments shared with others.
- Notifications alert business users when a report is updated, data is changed or the application is updated.
- Create your own mobile apps using the SAS SDK for iOS and SAS SDK for Android to create embedded insights:
- Personalize your mobile app with embedded SAS Visual Analytics content, your corporate logo and name of your choosing.
- Preconfigure your mobile app to connect to SAS servers and subscribe to specified reports.
- Develop completely customized mobile apps that embed SAS Visual Analytics content (e.g., GatherIQ).
- Manage and secure your mobile app and data by integrating with mobile device management (MDM) service (via new APIs).
- Embed full reports or individual objects in websites and web apps using the SAS Visual Analytics SDK:
- Combine insights from multiple reports in one location.
- User selections within an embedded SAS Visual Analytics object can drive other elements anywhere on the webpage.
- Geographical maps are enabled through Esri ArcGIS Online or OpenStreetMap.
- You can lasso data points on geographical maps to select specific data for further analysis.
- Geographical maps make it easy to visualize measurement variances over a geographical area.
- Access to all Esri basemaps and geosearch is available through Esri ArcGIS Online at no additional charge.
- Custom polygons (e.g., sales territories, voting districts, floor plans, seating charts) will let you see the world just as your business demands for it. These polygons can be animated to show how key metrics change over time.
- Geographic point clustering makes it easier to visualize high-volume location data and identify areas of interest. Get more or less details at different zoom levels.
- Add map pins to mark points of interest and insights on a map.
- With Esri ArcGIS Online license, you can enrich your data with Esri demographics data:
- Start from a pin, and select the area that can be traveled based on travel distance or provided travel time.
- Create travel routes between points.
- Understand how location affects outcomes by geocoding your data – adding latitude and longitude columns to your data based on location information in your data (country, state, zip code, city, street).
Security & administration
- SAS Environment Manager provides easy-to-use, web-based centralized administration and monitoring of your BI and analytics environment, including users, data, content, servers, services and security.
- User authentication and content authorization support governance.
- Object-level security (folders, reports, etc.) and data security (table and row level) support governance.
- Seamless integration with corporate identity directories such as LDAP.
- Rules-mapping application capabilities for users and groups support governance.
- Whitelist or blacklist mobile devices to determine authorization for SAS Visual Analytics apps.
- Near-real-time dashboard for monitoring system health and key activities.
- Distributed processing node addition and deletion.
- Scriptable APIs perform administrative tasks in batch, including management of security, libraries, users groups and configurations.
- Customizable monitoring and performance reports.
- Environmentwide log exploration, job scheduling and monitoring.
SAS® Viya® in-memory engine
- CAS (SAS Cloud Analytic Services) performs processing in memory and distributes processing across nodes in a cluster.
- User requests (expressed in a procedural language) are translated into actions with the parameters needed to process in a distributed environment. The result set and messages are passed back to the procedure for further action by the user.
- Data is managed in blocks and can be loaded in memory and on demand.
- If tables exceed memory capacity, the server caches the blocks on disk. Data and intermediate results are held in memory as long as required, across jobs and users.
- Includes highly efficient node-to-node communication. An algorithm determines the optimal number of nodes for a given job.
- Communication layer supports fault tolerance and lets you remove or add nodes from a server while it is running. All components can be replicated for high availability.
- Support for legacy SAS code and direct interoperability with SAS 9.4M5 clients.
- Supports multitenancy deployment, allowing for a shared software stack to support isolated tenants in a secure manner.
- On-site deployments:
- Single-machine server to support the needs of small to midsized organizations.
- Distributed server to meet growing data, increasing workloads and scalability requirements.
- Cloud deployments:
- Enterprise hosting.
- Private or public cloud (e.g., BYOL in Amazon) infrastructure.
- Cloud Foundry platform as a service (PaaS) to support multiple cloud providers.
Descriptive & predictive modeling
- Explore and evaluate segments for further analysis using k-means clustering, scatter plots and detailed summary statistics.
- Use machine learning techniques to build predictive models from a visual or programming interface.
View more descriptive & predictive modeling features