Find the value in your data, no matter the size – including big data stored in Hadoop. A drag-and-drop web browser interface empowers multiple users to explore massive data, then interactively and iteratively create descriptive and predictive models. Distributed, in-memory processing dramatically shortens model development time so you can run complex analytic computations in just minutes.
Beat competitors with precise insights.
Quickly surface insights hidden in vast stores of data. Discover, analyze and evaluate new opportunities that your competitors have missed. Find new ways to grow revenue. Powerful, predictive analytics with visual data exploration capabilities enable business analysts and statisticians do more with data than ever before.
Boost your analytcal teams' productivity.
Multiple users can quickly and interactively customize models – adding or changing variables, removing outliers, etc. – and instantaneously see how those changes affect model outcomes. Which model provides the most predictive power? It’s now easy to find out – and get more value from your big data analytics.
Run more models faster – with precision.
Our multicore processing environment reduces that to minutes. You can build models to target specific groups or segments, and run numerous multiple scenarios simultaneously. Ask more what-if questions, and get fast answers. Refined models produce better results.
Stay agile with in-memory computing.
Perform complex analytic computations using an in-memory engine. Modelers can quickly test new ideas, try different modeling techniques and refine models on the fly to produce the best results – using data volumes never before possible.
Demos & Screenshots
- Interactive data exploration and modeling environment. Quickly identify predictive drivers among multiple exploratory variables, and interactively discover outliers and data discrepancies through integration with SAS Visual Analytics Explorer. And create powerful descriptive and predictive models with a simple drag-and-drop interface.
- Descriptive modeling. Visually explore and evaluate segments for further analysis using k-means clustering, scatter plots and detailed summary statistics.
- Predictive modeling. Build predictive models using techniques such as linear regression, generalized linear modeling, logistic regression and classification trees.
- Dynamic group-by processing. Concurrently build models and process results for each group or segment without having to sort or index data each time.
- In-memory analytical processing. Build models faster on diverse sets of data, including Hadoop. There's no need to write data to disk or perform data shuffling, and you can instantly see the impact of changes (e.g., adding new variables or removing outliers).
- Model comparison and assessment. Generate model comparison summaries, such as lift charts, ROC charts, concordance statistics and misclassification tables on one or more models.
- Model scoring. Generate SAS DATA step code, and apply it to new data.
- Platform support. Supports Hadoop distributed file system (Cloudera or Hortonworks distributions), as well as Teradata, Greenplum (Pivotal) and Oracle databases.