Predictive Analytics

Organizations today need to mine historical data to gain new insights and form a reliable basis for accurate forecasting.

SQL Server 2012 Predictive Analytics provides a highly advanced data mining solution combined with the simplicity and familiarity of Excel. Use familiar tools and a comprehensive development environment for building sophisticated data mining solutions.

What's New in SQL Server 2012

  • Ability to test multiple data-mining models at once, while confirming stability.
  • Build multiple, incompatible models in a single structure.
  • Apply model analysis over filtered data.
  • Blend optimized near-term and stable long-term predictions.

Key Capabilities

  • Rely on a Complete Predictive Analytics Platform

    Make informed decisions with intuitive and comprehensive predictive insight available to all users.

    • Use rich and innovative algorithms like inventory forecast, most profitable customer identification, uncover non-intuitive data relationships and identify trends in unstructured data.
    • Enjoy a comprehensive development environment with SQL Server Development Tools environment.
    • Deliver insights with Microsoft Office by using familiar Excel environment with Data Mining add-in.
    • Take advantage of enterprise-class capabilities like high availability, superior performance, and scalability.
  • Integrate Predictive Analysis Everywhere

    Integrate prediction into every step of the data life cycle to discover hidden insights.

    • Extract quality data by using in-flight mining for ETL data loads.
    • Perform insightful analysis by including data-mining results as dimensions in OLAP cubes or add prediction functions to calculations and key performance indicators.
    • Natively integrate data reporting by using data-mining queries as the data source in Reporting Services.
    • Use predictive KPIs via integration of Analysis Services and SharePoint server and discovering attributes that influence KPIs.
  • Extend Predictive Analysis

    Extend prediction and enhance data-mining functionality to create intelligent applications.

    • Build applications by using predictive programming by take advantage of XMLA, Data Mining extensions, ADOMD.NET, OLE DB, and Analysis Management objects.
    • Customize algorithms and visualizations by using procedures, predictive model markup language, algorithms, and visualizations.

"Identifying trends and asking "what if" types of questions are the sort of things people will want to do going forward. Power View in SQL Server 2012 is quite a good exploratory tool for such purposes because it’s fast and very intuitive to use."

Jeremy Lane
Data Warehouse Manager, Anglia Support Partnership