Data Warehouse Analytics

Data Warehouse Analytics

DB2 can help you analyze data and explore information from different perspectives for a more extensive view of your business.

Cubing Services for OLAP

Online analytic processing (OLAP) allows you to interrogate data by navigating from summary to detail information. DB2 enables you to create, edit, import, export and deploy OLAP models in a data warehouse environment. It provides access to terabytes of data from multiple angles and perspectives so you can view and analyze it to make informed decisions.

DB2 supports dimensions, hierarchies, measures and summarizations as well as cross-dimensional calculations, time series and parallel period analyses. With dimensional navigation, you can slice, dice, drill and pivot data to gain insight into your business information.

Data Mining

Data mining enables you to analyze patterns and make predictions. It puts historical data through mathematical functions to determine business rules. Applying business rules to new data can help predict outcomes such as customer buying behaviors, up-sell or cross-sell opportunities, customer churn or fraud.

DB2 provides various data mining models and algorithms to support this process including segmentation, association, classification and regression.

  • Segmentation—offers deeper insight into information such as customer characteristics and helps you segment the data accordingly. For example, organizing customers into groups.
  • Association rules—can uncover patterns such as customer buying behaviour to help you gain more value from your data, for example, to increase cross and up-sell opportunities.
  • Sequential pattern rules—help you identify business rules based on event sequences such as what product a customer buys next or what leads to customer churn.
  • Fraud detection—allows you to find suspicious data records that could indicate misuse or fraud. For example, outlier rules can be applied to a banking transaction when it enters the system to help predict whether it is fraudulent.

Visualization enables visual analysis of data mining results, using a Java-based results browser. Java visualizers graphically present the results of associations, demographic clustering, sequences, numeric prediction and tree classification modeling operations. Visualizers also support data from other PMML-compliant data mining applications including SAS, SPSS and Angoss.

DB2 data mining capabilities allow you to:

  • Process data within the database, without the performance impact or risk of copying to an external system.
  • Predict and analyze patterns in a simpler, more intuitive way without complex setup or parameterization.
  • Integrate your solutions with different front-ends, infrastructures, databases and reporting tools (such as IBM Cognos) using an API.
  • Store, share and apply models across different applications.

Unstructured Data

Unstructured data represents up to 80 percent of the data within an organization. DB2 can extract information from your unstructured business text and correlate it with your structured data to increase business insight in areas such as customer attrition and product defect analysis.

DB2 provides drag and drop processing using two types of annotators: dictionary (word lookup and annotation) and pattern based extractors. You can also view your data in different ways to improve analysis such as by statistics or row and column information.

You can produce simple reports such as using text from customer surveys. The software also allows you to process unstructured data and create multidimensional reports using OLAP capabilities. In addition, unstructured data can be integrated into data mining models to broaden predictive capabilities.

Contact IBM

Considering a purchase?

DB2 Warehousing Resources