InfoSphere Warehouse

The data mining technologies from IBM help you to detect fraud, prevent customer churn, segment your customers, and simplify market basket analysis. The in-database data mining capabilities integrate with existing systems to provide scalable, high performing predictive and pattern analysis without moving your data into proprietary data mining platforms.

Unique features of InfoSphere Warehouse Data Mining

• Highly scalable, state-of-the-art data mining algorithms
  Profit from IBM’s state-of-the art technology, providing you with reliable, innovative algorithms for data analysis.
• Efficient and secure in-database processing
  Process your data directly in the database, without taking the effort and potential security risks of copying it to an external system.
• Easy to use even without in-depth data mining knowledge
  Apply data mining in a simple and intuitive way without complex setups or parameterization using the Easy Mining interfaces and the interactive flow creation of the Design Studio.
• Easy integration with other components
  Integrate your solutions with different front-ends, infrastructures, other databases and reporting tools (such as Cognos) through the powerful IBM Data Mining SQL API.
• Supports relevant standards
  Store, share and apply models across different applications using the Predictive Model Markup Language (PMML).

Modeling. The data mining process starts with historical data being gathered and put through a series of mathematical functions to derive business rules such as: "Customers buying Gouda cheese are likely to also buy mixed nuts 58% of the time." Business rules are collected together into a data mining model. Data mining models are represented using the Predictive Model Markup Language (PMML), which can be used to exchange mining models with other data mining tools like SAS or SPSS. Design Studio provides a convenient graphical environment for data modeling. For more information on supported data mining models, please visit here.

Visualization. Visualization of data mining models is an important functionality to evaluate the mining results. The powerful data mining visualization capabilities within InfoSphere Warehouse enable you to get fast insight into the discovered business rules. Read more

Scoring. The verified business rules can be applied to new data to determine the appropriate predicted outcome. For example, a new bank transaction enters the system and the fraud detection rules are applied against the data. The rules will predict the probability that the record is fraudulent. This process of applying the business rules is called Scoring. Scoring in real time allows businesses to catch fraudulent records faster, segment new customers and offer them better service, and detect defects quicker. InfoSphere Warehouse Data Mining performs Scoring directly in the database. This ensures a maximum of efficiency and security and allows you to embed Data Mining directly into your business processes.