Specifying Model Settings

Numerous configuration settings are available for configuring data mining models at build time. To specify settings, create a settings table with the columns shown in Table 5-7 and pass the table to CREATE_MODEL.

Table 5-7 Settings Table Required Columns

Column Name	Data Type
`setting_name`	`VARCHAR2(30)`
`setting_value`	`VARCHAR2(4000)`

Example 5-1 creates a settings table for an SVM classification model. Since SVM is not the default classifier, the ALGO_NAME setting is used to specify the algorithm. Setting the SVMS_KERNEL_FUNCTION to SVMS_LINEAR causes the model to be built with a linear kernel. If you do not specify the kernel function, the algorithm chooses the kernel based on the number of attributes in the data.

Some settings apply generally to the model, others are specific to an algorithm. Model settings are documented in Oracle Database PL/SQL Packages and Types Reference and referenced in Table 5-8 and Table 5-9.

Table 5-8 General Model Settings

Settings	Description
Mining function settings	See "Mining Function Settings" in Oracle Database PL/SQL Packages and Types Reference
Algorithm names	See "Algorithm Names" in Oracle Database PL/SQL Packages and Types Reference
Global model characteristics	See "Global Settings" in Oracle Database PL/SQL Packages and Types Reference
Automatic Data Preparation	See "Automatic Data Preparation" in Oracle Database PL/SQL Packages and Types Reference

Table 5-9 Algorithm-Specific Model Settings

Algorithm	Description
Decision Tree	See "Algorithm Settings: Decision Tree" in Oracle Database PL/SQL Packages and Types Reference
Expectation Maximization	See "Algorithm Settings: Expectation Maximization" in Oracle Database PL/SQL Packages and Types Reference
Generalized Linear Models	See "Algorithm Settings: Generalized Linear Models" in Oracle Database PL/SQL Packages and Types Reference
k-Means	See "Algorithm Settings: k-Means" in Oracle Database PL/SQL Packages and Types Reference
Naive Bayes	See "Algorithm Settings: Naive Bayes" in Oracle Database PL/SQL Packages and Types Reference
Non-Negative Matrix Factorization	See "Algorithm Settings: Non-Negative Matrix Factorization" in Oracle Database PL/SQL Packages and Types Reference
O-Cluster	See "Algorithm Settings: O-Cluster" in Oracle Database PL/SQL Packages and Types Reference
Singular Value Decomposition	See "Algorithm Settings: Singular Value Decomposition" in Oracle Database PL/SQL Packages and Types Reference
Support Vector Machine	See "Algorithm Settings: Support Vector Machine" in Oracle Database PL/SQL Packages and Types Reference

Example 5-1 Creating a Settings Table for an SVM Classification Model

CREATE TABLE svmc_sh_sample_settings (
  setting_name VARCHAR2(30),
  setting_value VARCHAR2(4000));

BEGIN 
  INSERT INTO svmc_sh_sample_settings (setting_name, setting_value) VALUES
    (dbms_data_mining.algo_name, dbms_data_mining.algo_support_vector_machines);
  INSERT INTO svmc_sh_sample_settings (setting_name, setting_value) VALUES
    (dbms_data_mining.svms_kernel_function, dbms_data_mining.svms_linear);
  COMMIT;
END;
/