Table of Contents

• List of Tables
• Title and Copyright Information
• Preface
  • Audience
  • Documentation Accessibility
  • Related Documentation
    • Oracle Data Mining Resources on the Oracle Technology Network
    • Application Development and Database Administration Documentation
  • Conventions
• Changes in This Release for Oracle Data Mining User's Guide
  • Oracle Data Mining User's Guide is New in This Release
  • Changes in Oracle Data Mining 12c Release 2 (12.2)
    • New Features in 12c Release 2
  • Changes in Oracle Data Mining 12c Release 1 (12.1)
    • New Features
    • Desupported Features
    • Other Changes
• 1 Data Mining With SQL
  • 1.1 Highlights of the Data Mining API
  • 1.2 Example: Targeting Likely Candidates for a Sales Promotion
  • 1.3 Example: Analyzing Preferred Customers
  • 1.4 Example: Segmenting Customer Data
  • 1.5 Example : Building an ESA Model with a Wiki Dataset
• 2 About the Data Mining API
  • 2.1 About Mining Models
  • 2.2 Data Mining Data Dictionary Views
    • 2.2.1 ALL_MINING_MODELS
    • 2.2.2 ALL_MINING_MODEL_ATTRIBUTES
    • 2.2.3 ALL_MINING_MODEL_PARTITIONS
    • 2.2.4 ALL_MINING_MODEL_SETTINGS
    • 2.2.5 ALL_MINING_MODEL_VIEWS
    • 2.2.6 ALL_MINING_MODEL_XFORMS
  • 2.3 Data Mining PL/SQL Packages
    • 2.3.1 DBMS_DATA_MINING
    • 2.3.2 DBMS_DATA_MINING_TRANSFORM
      • 2.3.2.1 Transformation Methods in DBMS_DATA_MINING_TRANSFORM
    • 2.3.3 DBMS_PREDICTIVE_ANALYTICS
  • 2.4 Data Mining SQL Scoring Functions
• 3 Preparing the Data
  • 3.1 Data Requirements
    • 3.1.1 Column Data Types
    • 3.1.2 Data Sets for Classification and Regression
    • 3.1.3 Scoring Requirements
  • 3.2 About Attributes
    • 3.2.1 Data Attributes and Model Attributes
    • 3.2.2 Target Attribute
    • 3.2.3 Numericals, Categoricals, and Unstructured Text
    • 3.2.4 Model Signature
    • 3.2.5 Scoping of Model Attribute Name
    • 3.2.6 Model Details
  • 3.3 Using Nested Data
    • 3.3.1 Nested Object Types
    • 3.3.2 Example: Transforming Transactional Data for Mining
  • 3.4 Using Market Basket Data
    • 3.4.1 Example: Creating a Nested Column for Market Basket Analysis
  • 3.5 Using Retail Analysis Data
    • 3.5.1 Example: Calculating Aggregates
  • 3.6 Handling Missing Values
    • 3.6.1 Examples: Missing Values or Sparse Data?
      • 3.6.1.1 Sparsity in a Sales Table
      • 3.6.1.2 Missing Values in a Table of Customer Data
    • 3.6.2 Missing Value Treatment in Oracle Data Mining
    • 3.6.3 Changing the Missing Value Treatment
• 4 Transforming the Data
  • 4.1 About Transformations
  • 4.2 Preparing the Case Table
    • 4.2.1 Creating Nested Columns
    • 4.2.2 Converting Column Data Types
    • 4.2.3 Text Transformation
    • 4.2.4 About Business and Domain-Sensitive Transformations
  • 4.3 Understanding Automatic Data Preparation
    • 4.3.1 Binning
    • 4.3.2 Normalization
    • 4.3.3 Outlier Treatment
    • 4.3.4 How ADP Transforms the Data
  • 4.4 Embedding Transformations in a Model
    • 4.4.1 Specifying Transformation Instructions for an Attribute
      • 4.4.1.1 Expression Records
      • 4.4.1.2 Attribute Specifications
    • 4.4.2 Building a Transformation List
      • 4.4.2.1 SET_TRANSFORM
      • 4.4.2.2 The STACK Interface
      • 4.4.2.3 GET_MODEL_TRANSFORMATIONS and GET_TRANSFORM_LIST
    • 4.4.3 Transformation Lists and Automatic Data Preparation
    • 4.4.4 Oracle Data Mining Transformation Routines
      • 4.4.4.1 Binning Routines
      • 4.4.4.2 Normalization Routines
      • 4.4.4.3 Routines for Outlier Treatment
  • 4.5 Understanding Reverse Transformations
• 5 Creating a Model
  • 5.1 Before Creating a Model
  • 5.2 The CREATE_MODEL Procedure
    • 5.2.1 Choosing the Mining Function
    • 5.2.2 Choosing the Algorithm
    • 5.2.3 Supplying Transformations
      • 5.2.3.1 Creating a Transformation List
      • 5.2.3.2 Transformation List and Automatic Data Preparation
    • 5.2.4 About Partitioned Model
      • 5.2.4.1 Partitioned Model Build Process
      • 5.2.4.2 DDL in Partitioned model
        • 5.2.4.2.1 Drop Model or Drop Partition
        • 5.2.4.2.2 Add Partition
      • 5.2.4.3 Partitioned Model scoring
  • 5.3 Specifying Model Settings
    • 5.3.1 Specifying Costs
    • 5.3.2 Specifying Prior Probabilities
    • 5.3.3 Specifying Class Weights
    • 5.3.4 Model Settings in the Data Dictionary
    • 5.3.5 Specifying Mining Model Settings for R Model
      • 5.3.5.1 ALGO_EXTENSIBLE_LANG
      • 5.3.5.2 RALG_BUILD_FUNCTION
        • 5.3.5.2.1 RALG_BUILD_PARAMETER
      • 5.3.5.3 RALG_DETAILS_FUNCTION
        • 5.3.5.3.1 RALG_DETAILS_FORMAT
      • 5.3.5.4 RALG_SCORE_FUNCTION
      • 5.3.5.5 RALG_WEIGHT_FUNCTION
      • 5.3.5.6 Registered R Scripts
      • 5.3.5.7 R Model Demonstration Scripts
  • 5.4 Model Detail Views
    • 5.4.1 Model Detail Views for Association Rules
    • 5.4.2 Model Detail View for Frequent Itemsets
    • 5.4.3 Model Detail View for Transactional Itemsets
    • 5.4.4 Model Detail View for Transactional Rule
    • 5.4.5 Model Detail Views for Classification Algorithms
    • 5.4.6 Model Detail Views for Decision Tree
    • 5.4.7 Model Detail Views for Generalized Linear Model
    • 5.4.8 Model Detail Views for Naive Bayes
    • 5.4.9 Model Detail View for Support Vector Machine
    • 5.4.10 Model Detail Views for Clustering Algorithms
    • 5.4.11 Model Detail Views for Expectation Maximization
    • 5.4.12 Model Detail Views for k-Means
    • 5.4.13 Model Detail Views for O-Cluster
    • 5.4.14 Model Detail Views for Explicit Semantic Analysis
    • 5.4.15 Model Detail Views for Non-Negative Matrix Factorization
    • 5.4.16 Model Detail Views for Singular Value Decomposition
    • 5.4.17 Model Detail View for Minimum Description Length
    • 5.4.18 Model Detail View for Binning
    • 5.4.19 Model Detail Views for Global Information
    • 5.4.20 Model Detail View for Normalization and Missing Value Handling
• 6 Scoring and Deployment
  • 6.1 About Scoring and Deployment
  • 6.2 Using the Data Mining SQL Functions
    • 6.2.1 Choosing the Predictors
    • 6.2.2 Single-Record Scoring
  • 6.3 Prediction Details
    • 6.3.1 Cluster Details
    • 6.3.2 Feature Details
    • 6.3.3 Prediction Details
    • 6.3.4 GROUPING Hint
  • 6.4 Real-Time Scoring
  • 6.5 Dynamic Scoring
  • 6.6 Cost-Sensitive Decision Making
  • 6.7 DBMS_DATA_MINING.Apply
• 7 Mining Unstructured Text
  • 7.1 About Unstructured Text
  • 7.2 About Text Mining and Oracle Text
  • 7.3 Data Preparation for Text Features
  • 7.4 Creating a Model that Includes Text Mining
  • 7.5 Creating a Text Policy
  • 7.6 Configuring a Text Attribute
• 8 Administrative Tasks for Oracle Data Mining
  • 8.1 Installing and Configuring a Database for Data Mining
    • 8.1.1 About Installation
    • 8.1.2 Enabling or Disabling a Database Option
    • 8.1.3 Database Tuning Considerations for Data Mining
  • 8.2 Upgrading or Downgrading Oracle Data Mining
    • 8.2.1 Pre-Upgrade Steps
      • 8.2.1.1 Dropping Models Created in Java
      • 8.2.1.2 Dropping Mining Activities Created in Oracle Data Miner Classic
    • 8.2.2 Upgrading Oracle Data Mining
      • 8.2.2.1 Using Database Upgrade Assistant to Upgrade Oracle Data Mining
        • 8.2.2.1.1 Upgrading from Release 10g
        • 8.2.2.1.2 Upgrading from Release 11g
      • 8.2.2.2 Using Export/Import to Upgrade Data Mining Models
        • 8.2.2.2.1 Export/Import Release 10g Data Mining Models
        • 8.2.2.2.2 Export/Import Release 11g Data Mining Models
    • 8.2.3 Post Upgrade Steps
    • 8.2.4 Downgrading Oracle Data Mining
  • 8.3 Exporting and Importing Mining Models
    • 8.3.1 About Oracle Data Pump
    • 8.3.2 Options for Exporting and Importing Mining Models
    • 8.3.3 Directory Objects for EXPORT_MODEL and IMPORT_MODEL
    • 8.3.4 Using EXPORT_MODEL and IMPORT_MODEL
    • 8.3.5 Importing From PMML
  • 8.4 Controlling Access to Mining Models and Data
    • 8.4.1 Creating a Data Mining User
      • 8.4.1.1 Granting Privileges for Data Mining
    • 8.4.2 System Privileges for Data Mining
    • 8.4.3 Object Privileges for Mining Models
  • 8.5 Auditing and Adding Comments to Mining Models
    • 8.5.1 Adding a Comment to a Mining Model
    • 8.5.2 Auditing Mining Models
• A The Data Mining Sample Programs
  • A.1 About the Data Mining Sample Programs
  • A.2 Installing the Data Mining Sample Programs
  • A.3 The Data Mining Sample Data
• Index