• Nenhum resultado encontrado

Oracle® Data Mining

N/A
N/A
Protected

Academic year: 2023

Share "Oracle® Data Mining"

Copied!
84
0
0

Texto

This manual explains how to install Oracle Data Mining and prepare the database for mining activities. The Oracle Data Mining documentation set is part of the Oracle Database 11g Release 2 (11.2) Online Documentation Library. For detailed information about the Oracle Data Mining PL/SQL interface, see the Oracle Database PL/SQL Packages and Types Reference.

The Oracle Data Mining documentation set does not cover the topic of customizing a database for data mining. See: Oracle Data Mining Concepts for a complete list of new and deprecated features in Oracle Data Mining 11g. Oracle Data Mining 11g is more tightly integrated with Oracle Database than in previous releases.

The Database Configuration Assistant (DBCA) is no longer required to add Oracle Data Mining metadata to a database. The Oracle Data Mining Scoring Engine, a separately installable database option in Oracle 10g, is not available in Oracle 11g.

Step One: Install the Database

To quickly install the Oracle Data Mining software and prepare to start mining your data, you can follow the steps described in this chapter. These steps explain how to install Oracle Data Mining locally on your Windows PC or laptop and launch the client interfaces: Oracle Data Miner and Oracle Spreadsheet Add-In for Predictive Analytics.

Step Two: Verify the Data Mining Installation

Step Three: Create a Data Mining User

If the user needs to import or export data mining models, grant this additional right.

Step Four: Install and Launch Oracle Data Miner

Step Five: Install and Launch the Spreadsheet Add-In

Step Six: If You Want to Use the Sample Programs

This chapter is intended for anyone who wants to install Oracle Data Mining on a personal computer or laptop running Microsoft Windows.

Install Oracle Database

Before you begin the installation, ensure that your computer meets the system requirements described in Oracle Database Installation Guide for Microsoft Windows. Note: To stop an Oracle database, first stop the database service (OracleServiceservice_name), then stop the database listener (OracleOraDb11g_homeTNSListener). This will cause the installer to launch Oracle Database Configuration Assistant once the software is installed.

The Global Database name is the fully qualified name of the database: for example, dbname.us.oracle.com. The installer analyzes your computer to determine if you have the necessary resources for the installation. After the software has been copied to your computer, the Database Configuration Assistant creates and starts the initial database.

Select Password Management if you want to reset the administrative passwords or unlock any other accounts provided with the database. Note: If you want to use Oracle sample data or the Oracle Data Mining sample programs, unlock the SH account.

Verify the Data Mining Installation

Install Oracle Database Examples

Ensure that your computer meets the system requirements described in Oracle Database Examples Installation Guide. The installer performs prerequisite checks to determine if you have the resources required for the installation.

Install Oracle Data Miner

Data Miner: Requirements

Data Miner: Installation Steps

Start Oracle Data Miner

Install the Spreadsheet Add–In

Spreadsheet Add–In: Requirements

Spreadsheet Add–In: Install Steps

Although the add-in can use data stored in Excel or Oracle Database, a connection to Oracle Database is required.

Start Data Mining in Excel

Connecting to an Oracle Database

Connect Oracle Data Miner

Connect Application Development Tools

Connect SQL*Plus

Connect SQL Developer

For Port, specify the port on which the database will listen for requests from the client. This is the first component of the global database name described in step 6 of the installation instructions in Chapter 2.

Connect Using Java

Oracle Client and Oracle Net

Install Oracle Client

Ensure that your computer meets the system requirements described in the Oracle Database Client Installation Guide for Microsoft Windows. If you are unsure whether your user ID is in the Administrators group, check User Accounts in the Windows Control Panel. To start the installation, go to the client installation directory and run SETUP.EXE to start the Oracle Universal Installer.

On the Select Product Language page, select the language for the Oracle Client user interface. Note: The following instructions are not intended as a substitute for Oracle Client and Net Management documentation. On the Specify Installation Location page, specify the path to the base directory where you installed Oracle Database.

This will be the home directory of the Oracle client you are installing (Oracle home). The Oracle Net Configuration Assistant determines that the Oracle Client can use the Easy Connect naming method, as described in "Oracle Client and Oracle Net" on page 3-4. Click the Manage Passwords button at the bottom of the page to unlock any schemes you plan.

If you are going to use the Oracle Data Mining sample programs and you did not unlock the SH schema when you installed the database, you should do so now. Click Exit to exit Oracle Net Configuration Assistant, then click Finish to complete the installation of Oracle Client. If you stopped any Oracle services before using Oracle Universal Installer, you can restart them at this time.

Create a Net Service Name

Specify the user ID and password of an account with data mining privileges on the host. If the username and password were the only problem with the connection, Oracle Net Configuration Assistant makes a successful test connection.

Connect the Spreadsheet Add–In

Users and Privileges for Data Mining

Create a Database User

Grant Data Mining Privileges

System Privileges for Data Mining

Object Privileges for Mining Models

In this chapter, you will learn how to find information about mining models in the data dictionary and how to perform various operations on mining models.

Obtaining Information from the Data Dictionary

Obtaining Information about Mining Models

Obtaining Information about Mining Model Attributes

Obtaining Information about Mining Model Settings

Adding a Comment to a Mining Model

Auditing Mining Models

Enabling Auditing in the Database

Opening an Audit Trail on Mining Models

Closing the Audit Trail

Viewing the Audit Trail

Exporting and Importing Mining Models

When a DBA exports an entire database using EXPDP, all data mining models in the database are exported. When a DBA or individual user exports a schema using EXPDP, all data mining models in the schema are exported. You can select all mining models in a schedule or mining models that meet specific criteria.

The Data Pump export utility writes the tables and metadata that make up a model to a set of dump files, which consists of one or more files. Data Pump's import program takes the tables and metadata from the dump file and restores them to the target database.

Prerequisites

Directory Objects

For more information about creating database directories, see the CREATE DIRECTORY and GRANT commands in the Oracle Database SQL Language Reference.

Additional Database Privileges

PL/SQL APIs for Exporting and Importing Models

Java APIs for Exporting and Importing Models

Exporting Models from the DMUSER Schema

Importing Models Into the Same Schema

Importing Models Into a Different Schema

Handles object type TABLE_EXPORT/TABLE/GRANT/OWNER_GRANT/OBJECT_GRANT Handles object type TABLE_EXPORT/TABLE/INDEX/INDEX. Handles object type TABLE_EXPORT/TABLE/INDEX/STATISTICS/INDEX_STATISTICS Handles object type TABLE_EXPORT/TABLE/STATISTICS/TABLE_STATISTICS Handles object type TABLE_EXPORT/TABLE/POST_INSTANCE/PROCACT_INSTANCE. The job "SYS12" 1 was completed at 12"1: "1 successfully at

Exporting and Importing Specific Models

Upgrading Oracle Data Mining

Data Mining Upgrade Overview

Data Mining Model Upgrade

Upgrading from 10g

Upgrading from 11.1

Upgrade Using Database Upgrade Assistant

Upgrade Using Export/Import

Export/Import Release 11.1 Data Mining Models

Export/Import Release 10g Data Mining Models

Shutting down the database before running on the upgraded mining models will also empty the shared pool. After the import is complete and there is no need to downgrade, you must remove the DMSYS schema from the database.

Post Upgrade Steps

Downgrade Data Mining

You can learn a lot about the Oracle Data Mining APIs from the Data Mining sample programs. The programs illustrate typical approaches to data preparation, algorithm selection, algorithm tuning, testing, and scoring.

Installation and Setup

Install the Sample Programs

If database samples are not installed, you can perform the installation by following the instructions in "Installing Oracle Database Samples" on page 2-8.

Run the Configuration Scripts

Locate the Sample Programs

Run the Sample Programs

Run the PL/SQL Sample Programs

Prepare to Run the Java Programs

Voeg ORACLE_BASE\ORACLE_HOME\jdk\bin\ by jou PATH-veranderlike voor die paaie van enige ander Java-weergawes. ORACLE_BASE\ORACLE_HOME\RDBMS\jlib\jdm.jar ORACLE_BASE\ORACLE_HOME\RDBMS\jlib\ojdm_api.jar ORACLE_BASE\ORACLE_HOME\RDBMS\jlib\xdb.jar ORACLE_BASE\\ORACLE_bjd5.jar.jar.jar. ORACLE_BASE\ORACLE_HOME\oc4j\j2ee\home\lib\connector.jar ORACLE_BASE\ORACLE_HOME\jlib\orai18n.jar.

ORACLE_BASE\ORACLE_HOME\jlib\orai18n-mapping.jar ORACLE_BASE\ORACLE_HOME\lib\xmlparserv2.jar. To use the JAVAC executable, open a command window and navigate to \RDBMS\demo in the Oracle home.

Run the Java Programs

List the Sample Models

The PL/SQL Programs

PL/SQL Programs: Algorithms

PL/SQL Programs: Mining Functions

PL/SQL Text Mining Programs

The Java Programs

Java Programs: Algorithms

Java Programs: Mining Functions

Java Text Mining Programs

The Sample Data

Customer Data for Data Mining

SELECT a.CUST_ID, a.CUST_GENDER, 2003-a.CUST_YEAR_OF_BIRTH AGE, a.CUST_MARITAL_STATUS, c.COUNTRY_NAME, a.CUST_INCOME_LEVEL, b.EDUCATION, b.OCCUPATION, b.HOUSEHOLD_SIZE, b.YRS_RESIDENCE, b.AFFINITY_CARD, b. BULK_PACK_DISKETTES, b.FLAT_PANEL_MONITOR, b.HOME_THEATER_PACKAGE, b.BOOKKEEPING_APPLICATION, . Each view has a CUSTOMER_ID column, the case ID, and an AFFINITY_CARD column, the target used by the predictive models. CUST_ID NOT NULL NUMBER CUST_GENDER NOT NULL CHAR(1) AGE NUMBER CUST_MARITAL_STATUS VARCHAR2(20) COUNTRY_NAME NOT NULL VARCHAR2(40) CUST_INCOME_LEVEL VARCHAR2(30) EDUCATION VARCHAR2(21) PROFESSIONAL VARCHAR2(21) HOUSEHOLD_SIZE VARCHAR2(21) YRS_RES IDENTITY NUMBER AFFINITY_CARD NUMBER( 10 ) BULK_PACK_DISKETTES NUMBER(10) FLAT_PANEL_MONITOR NUMBER(10) HOME_THEATER_PACKAGE NUMBER(10) BOOKKEEPING_APPLICATION NUMBER(10) PRINTER_SUPPLIES NUMBER(10) Y_BOX_GAMES NUMBER(10) OS_DOC_SET_KANJI NUMBER(10).

Market Basket Data for Association Rules

Customer Data for Text Mining

In the MINING_*_NESTED_TEXT tables, the COMMENTS column is of type DM_NESTED_NUMERICALS. This plugin provides a set of SQL and PL/SQL commands that test whether or not basic data mining capabilities are present in the database. You can run these commands directly on the database without downloading or installing any additional software.

Connect to the Database

Create a Script or Execute the Commands Individually

Index

Imagem

Table Name Description

Referências

Documentos relacionados

This continuous process contains three steps: a production step using microbial mixed culture, a PHA accumulation step, and a PHA extraction step using NaClO (5%) disruption for 24