Skip Headers
Oracle® Data Mining Concepts
11g Release 1 (11.1)

Part Number B28129-01
Go to Documentation Home
Go to Book List
Book List
Go to Table of Contents
Go to Index
Go to Master Index
Master Index
Go to Feedback page
Contact Us

Go to previous page
Go to next page
View PDF

3 Introducing Oracle Predictive Analytics

This chapter presents an overview of Oracle Data Mining predictive analytics, an automated form of predictive data mining.

See Also:

This chapter includes the following sections:

About Predictive Analytics

Predictive Analytics is a technology that captures data mining processes in simple routines. Sometimes called "one-click data mining," predictive analytics simplifies and automates the data mining process.

Predictive analytics develops profiles, discovers the factors that lead to certain outcomes, predicts the most likely outcomes, and identifies a degree of confidence in the predictions.

Predictive Analytics and Data Mining

Predictive analytics uses data mining technology, but knowledge of data mining is not needed to use predictive analytics.

You can use predictive analytics simply by specifying an operation to perform on your data. You do not need to create or use mining models or understand the mining functions and algorithms summarized in Chapter 2 of this manual.

How Does it Work?

The predictive analytics routines analyze the input data and create mining models. These models are trained and tested and then used to generate the results returned to the user. The models and supporting objects are not preserved after the operation completes.

When you use data mining technology directly, you create a model or use a model created by someone else. You apply the model to new data (different from the data used to train and test the model). Predictive analytics routines apply the model to the same data used for training and testing.

See Also:

"Behind the Scenes" to gain insight into the inner workings of Oracle predictive analytic

Predictive Analytics Operations

Oracle Data Mining predictive analytics operations are described in Table 3-1.

Table 3-1 Oracle Predictive Analytics Operations

Operation Description


Explains how the individual attributes affect the variation of values in a target column


For each case, predicts the values in a target column


Creates a set of rules for cases that imply the same target value

Oracle Spreadsheet Add-In for Predictive Analytics

The Oracle Spreadsheet Add-In for Predictive Analytics provides predictive analytics operations within a Microsoft Excel spreadsheet. You can analyze Excel data or data that resides in an Oracle database.

Figure 3-1 shows the EXPLAIN operation using Microsoft Excel 7.0.

Figure 3-1 EXPLAIN in Oracle Spreadsheet Add-In for Predictive Analytics

Description of Figure 3-1 follows
Description of "Figure 3-1 EXPLAIN in Oracle Spreadsheet Add-In for Predictive Analytics"

Figure 3-2 shows the PREDICT operation.

Figure 3-2 PREDICT in Oracle Spreadsheet Add-In for Predictive Analytics

Description of Figure 3-2 follows
Description of "Figure 3-2 PREDICT in Oracle Spreadsheet Add-In for Predictive Analytics"

Figure 3-3 shows the PROFILE operation.

Figure 3-3 PROFILE in Oracle Spreadsheet Add-In for Predictive Analytics

Description of Figure 3-3 follows
Description of "Figure 3-3 PROFILE in Oracle Spreadsheet Add-In for Predictive Analytics"

You can download the latest version of the Spreadsheet Add-In from the Oracle Technology Network.

APIs for Predictive Analytics

Oracle Data Mining implements predictive analytics in the PL/SQL and Java APIs.

Predictive Analytics in the PL/SQL API

DBMS_PREDICTIVE_ANALYTICS package. The following SQL DESCRIBE shows the predictive analytics procedures with their parameters.

SQL> describe dbms_predictive_analytics

 Argument Name                  Type                    In/Out Default?
 ------------------------------ ----------------------- ------ --------
 DATA_TABLE_NAME                VARCHAR2                IN
 EXPLAIN_COLUMN_NAME            VARCHAR2                IN
 RESULT_TABLE_NAME              VARCHAR2                IN
 DATA_SCHEMA_NAME               VARCHAR2                IN     DEFAULT

 Argument Name                  Type                    In/Out Default?
 ------------------------------ ----------------------- ------ --------
 ACCURACY                       NUMBER                  OUT
 DATA_TABLE_NAME                VARCHAR2                IN
 CASE_ID_COLUMN_NAME            VARCHAR2                IN
 TARGET_COLUMN_NAME             VARCHAR2                IN
 RESULT_TABLE_NAME              VARCHAR2                IN
 DATA_SCHEMA_NAME               VARCHAR2                IN     DEFAULT

 Argument Name                  Type                    In/Out Default?
 ------------------------------ ----------------------- ------ --------
 DATA_TABLE_NAME                VARCHAR2                IN
 TARGET_COLUMN_NAME             VARCHAR2                IN
 RESULT_TABLE_NAME              VARCHAR2                IN
 DATA_SCHEMA_NAME               VARCHAR2                IN     DEFAULT

Predictive Analytics in the Java API

The Oracle Data Mining Java API defines predictive analytics tasks that use the DBMS_PREDICTIVE_ANALYTICS package.

The create methods that are defined under oracle.dmt.jdm.task.OraPredictiveAnalyticsTaskFactory class are shown as follows:

public OraPredictTask createPredictTask(String inputDataURI, String caseID,
                                        String targetColumn,
                                        String predictionResultTableName) 
                                 throws JDMException;
public OraExplainTask createExplainTask(String inputDataURI,
                                        String explainColumn,
                                        String explainResultTableName) 
                                 throws JDMException;
public OraProfileTask createProfileTask(String inputDataURI,
                                        String targetAttributeName,
                                        String profileResultTableName) 
                                 throws JDMException;

Example: Use OraProfileTask to Create Profile Results

//Step-1: Create OraPredictiveAnalyticsTaskFactory
OraPredictiveAnalyticsTaskFactory m_paFactory =

//Step-2: Create, save and execute OraProfileTask. After
OraProfileTask m_profileTask = 
   m_paFactory.createProfileTask("MINING_DATA_BUILD_V", "AFFINITY_CARD", 
  m_dmeConn.saveObject("profileTask_jdm", m_profileTask, true);

Example: PREDICT

Example 3-1 shows how a simple PREDICT operation can be used to find the customers most likely to increase spending if given an affinity card.

The customer data, including current affinity card usage and other information such as gender, education, age, and household size, is stored in a view called MINING_DATA_APPLY_V. The results of the PREDICT operation are written to a table named p_result_tbl.

The PREDICT operation calculates both the prediction and the accuracy of the prediction. Accuracy, also known as predictive confidence, is a measure of the improvement over predictions that would be generated by a naive model. In the case of classification, a naive model would always guess the most common class. In Example 3-1, the improvement is almost 50%.

Example 3-1 Predict Customers Most Likely to Increase Spending with an Affinity Card

p_accuracy NUMBER(10,9);
       accuracy                => p_accuracy,
       data_table_name         =>'mining_data_apply_v',
       case_id_column_name     =>'cust_id',
       target_column_name      =>'affinity_card',
       result_table_name       =>'p_result_tbl');
  DBMS_OUTPUT.PUT_LINE('Accuracy: ' || p_accuracy);

Accuracy: .492433267

The following query returns the gender and average age of customers most likely to respond favorably to an affinity card.

SELECT cust_gender, COUNT(*) as cnt, ROUND(AVG(age)) as avg_age
             FROM mining_data_apply_v a, p_result_tbl b
      WHERE a.cust_id = b.cust_id
         AND b.prediction = 1
      GROUP BY a.cust_gender
      ORDER BY a.cust_gender; 
C        CNT    AVG_AGE
- ---------- ----------
F         90         45
M        443         45

Behind the Scenes

This section provides some high-level information about the inner workings of Oracle predictive analytics. If you know something about data mining, you will find this information to be straight-forward and easy to understand. If you are unfamiliar with data mining, you can skip this section. You do not need to know this information to use predictive analytics.

See Also:

Chapter 2 for an overview of model functions and algorithms


EXPLAIN creates an attribute importance model. Attribute importance uses the Minimum Description Length algorithm to determine the relative importance of attributes in predicting a target value. EXPLAIN returns a list of attributes ranked in relative order of their impact on the prediction. This information is derived from the model details for the attribute importance model.

Attribute importance models are not scored against new data. They simply return information (model details) about the data you provide.

Attribute importance is described in "Attribute Importance".


PREDICT creates a Support Vector Machine (SVM) model for classification or regression.

PREDICT creates a Receiver Operating Characteristics (ROC) curve to analyze the per-case accuracy of the predictions. PREDICT optimizes the probability threshold for binary classification models. The probability threshold is the probability that the model uses to make a positive prediction. The default is 50%.


PREDICT returns a value indicating the accuracy, or predictive confidence, of the prediction. The accuracy is the improvement gained over a naive prediction. For a categorical target, a naive prediction would be the most common class, for a numerical target it would be the mean. For example, if a categorical target can have values small, medium, or large, and small is predicted more often than medium or large, a naive model would return small for all cases. Predictive analytics uses the accuracy of a naive model as the baseline accuracy.

The accuracy metric returned by PREDICT is a measure of improved maximum average accuracy versus a naive model's maximum average accuracy. Maximum average accuracy is the average per-class accuracy achieved at a specific probability threshold that is greater than the accuracy achieved at all other possible thresholds.

SVM is described in Chapter 18.


PROFILE creates a Decision Tree model to identify the characteristics of the attributes that predict a common target. For example, if the data has a categorical target with values small, medium, or large, PROFILE would describe how certain attributes typically predict each size.

The Decision Tree algorithm creates rules that describe the decisions that affect the prediction. The rules, expressed in XML as if-then-else statements, are returned in the model details. PROFILE returns XML that is derived from the model details generated by the algorithm.

Decision Tree is described in Chapter 11.