Enhance Applications with Predictions and Insights

Oracle9i Data Mining enables companies to systematize the extraction and integration of new business intelligence within their operations. Application developers can use Oracle9i Data Mining's Java- based API to add data mining insights and predictions to enhance business applications such as Customer Relationship Management (CRM), Enterprise Resource Planning (ERP), Web portals, and wireless applications. Telecommunications companies, for example, can use Oracle9i Data Mining to build churn applications that identify customers that are likely to churn before they leave for a competitor. Oracle9i Data Mining's predictions can be used to anticipate customer behavior and proactively manage them in mutually beneficial 1-to-1 relationships.

Retailers and database marketers can use Oracle9i Data Mining to build marketing campaign applications that target those prospects that are most likely to respond to offers. Oracle9i Data Mining can integrate data mining results into applications. Examples include predicting a customer's likelihood to churn, respond to a special offer, be a profitable customer, file a claim, or spend large amounts of money. E-businesses and Web sites can enhance Web searches using Oracle9i Data Mining to present other documents or items that are related or "associated" in use or content.

Once the data has been mined and the predictive models built, Oracle9i Data Mining can apply the models to "score" other data to make predictions. Scoring of data occurs in the database and scores are then available for use by other applications. Data mining models stored in the database can provide insights and predictions on demand to interactive applications, such as call centers, that suggest "recommendations." For example, a call center application could use a customer's historical data together with responses from a call in progress to rate the customer's preferences and make personalized cross-sell recommendations.

Prediction and Classification

Oracle9i Data Mining provides the Naïve Bayes data mining algorithm for making predictions and classifications. This algorithm is applicable to a variety of data mining problems and provides high accuracy. By finding patterns in data, companies can make predictions about the future behavior of customers with similar characteristics — using the past as a predictor of the future. Typical prediction applications estimate the probability of an outcome, such as "0, 1" or "yes, no" or "A, B, C or D." Consider the following example:

Question: Will this customer respond to my special offer?

Answer: "Yes," with a likelihood of 92%.

Oracle9i Data Mining's predictive models return predicted
outcomes and their associated probability, so companies
can proactively manage their business.

Oracle9i Data Mining's predictive models return predicted outcomes and their associated probability, so companies can proactively manage their business.

Results of models can be combined to provide valuable business intelligence. For example, Oracle9i Data Mining could build a model to predict the lifetime value (LTV) of a customer and another model to predict the likelihood that a customer will churn. Multiplying the probabilities of the two predicted outcomes (P(LTV) x P(Churn)) can provide valuable insights on how to spend your marketing budget.

Oracle9i Data Mining's predictions and classifications can be examined
using other software and applications, such as Oracle Discoverer shown here.

Finding Associations

Oracle9i Data Mining provides the Association Rules data mining algorithm to detect "associated" or co-occurring events hidden in databases. Association analysis is often used to find popular product bundles (e.g. market basket analysis) of products that are related for customers, such as "milk" and "cereal" being associated with "bananas." Associations can also be used to identify co-occurring items or events such as:

• What manufactured parts and equipment settings are associated with failure events?
• What patient and drug attributes are associated with which outcomes?
• Which items or products is this person most likely to buy or like?

Associations can be used to predict the next item placed into the shopping basket which can be helpful to satisfy customers and increase average order value.

Oracle Discoverer showing the results of Oracle9i Data Mining's association analysis.

Application developers access Oracle9i Data Mining's functionality through a Java-based API. Programmatic control of all data mining functions enables automation of data preparation, model building, and model scoring operations.

Java Data Mining (JDM) is an emerging data mining standard, following SUN's Java Community Process as a Java Specification Request (JSR). JDM has participation from Oracle, Sun, IBM, and many other companies that recognize the need for a Java-based standard for specifying and using data mining. JDM leverages several evolving data mining standards, including Object Management Group's Common Warehouse Metadata (CWM), the Data Mining Group's Predictive Mining Markup Language (PMML), and International Standards Organization's SQL/MM for Data Mining.