1
Introduction to Oracle Data Mining

This chapter describes what data mining is, what Oracle Data Mining is, and outlines the data mining process.

1.1 What is Data Mining?

Too much data and not enough information -- this is a problem facing many businesses and industries.

A solution lies here, with data mining. Most businesses have an enormous amount of data, with a great deal of information hiding within it, but "hiding" is usually exactly what it is doing: So much data exists that it overwhelms traditional methods of data analysis.

Data mining provides a way to get at the information buried in the data. Data mining finds hidden patterns in large, complex collections of data, patterns that elude traditional statistical approaches to analysis.

1.2 What Is Oracle Data Mining?

Oracle Data Mining (ODM) embeds data mining within the Oracle database. There is no need to move data out of the database into files for analysis and then back from files into the database for storing. The data never leaves the database -- the data, data preparation, model building, and model scoring results all remain in the database. This enables Oracle to provide an infrastructure for application developers to integrate data mining seamlessly with database applications.

ODM is designed to support production data mining in the Oracle database. Production data mining is most appropriate for creating applications to solve problems such as customer relationship management, churn, etc., that is, any data mining problem for which you want to develop an application.

ODM provides single-user milt-session access to models. Model building is either synchronous in the PL/SQL interface or asynchronous in the Java interface.

1.2.1 Oracle Data Mining Programming Interfaces

ODM integrates data mining with the Oracle data base and exposes data mining through the following interfaces:

• Java interface: Allows users to embed data mining in Java applications.
• DBMS_DATA_MINING and DBMS_DATA_MINING_TRANSFORM: Allow users to embed data mining in PL/SQL applications.

  Note: The Java and PL/SQL interfaces do not produce models that are interoperable. For example, you cannot produce a model with Java and apply it using PL/SQL, or vice versa, in this release.

The ODM Java interface and DBMS_DATA_MINING have similar, but not identical, capabilities. For a comparison of the interfaces, see Appendix A.

1.2.2 ODM Data Mining Functions

Data mining functions are based on two kinds of learning: supervised (directed) and unsupervised (undirected).

Supervised learning functions are typically used to predict a value, and are sometimes referred to as predictive models. Unsupervised learning functions are typically used to find the intrinsic structure, relations, or affinities in data but no classes or labels are assigned aprioi. These are sometimes referred to as descriptive models.

Oracle Data Mining supports the following data mining functions:

• Predictive models (supervised learning):
  • Classification: grouping items into discrete classes and predicting which class an item belongs to
  • Regression: function approximation and forecast of continuous values
  • Attribute importance: identifying the attributes that are most important in predicting results (Java interface only)
• Descriptive models (unsupervised learning):
  • Clustering: finding natural groupings in the data
  • Association models: "market basket" analysis
  • Feature extraction: create new attributes (features) as a combination of the original attributes
• Multimedia (TEXT)
• Bioinformatics (BLAST)