Contents

List of Figures

List of Tables

Title and Copyright Information

Preface

• Intended Audience
• Documentation Accessibility
• Related Documents
• Conventions

1 Introduction to Oracle Data Mining

• 1.1 What is Data Mining?
• 1.2 What Is Data Mining in the Database?
• 1.3 What Is Oracle Data Mining?
• • 1.3.1 Data Mining Functions
• 1.4 New Features

2 Data for Oracle Data Mining

• 2.1 Data, Cases, and Attributes
• 2.2 Data Requirements
• • 2.2.1 ODM Data Table Format
  • 2.2.2 Column Data Types Supported by ODM
  • • 2.2.2.1 Nested Columns in ODM
  • 2.2.3 Missing Values
  • • 2.2.3.1 Missing Values and NULL Values in ODM
    • 2.2.3.2 Missing Value Handling
  • 2.2.4 Sparse Data
  • 2.2.5 Outliers and Oracle Data Mining
• 2.3 Data Preparation
• • 2.3.1 Winsorizing and Trimming
  • 2.3.2 Binning (Discretization)
  • • 2.3.2.1 Methods for Computing Bin Boundaries
  • 2.3.3 Normalization

3 Supervised Data Mining

• 3.1 Classification
• • 3.1.1 Algorithms for Classification
  • • 3.1.1.1 Decision Tree Algorithm
    • 3.1.1.2 Naive Bayes Algorithm
    • 3.1.1.3 Adaptive Bayes Network Algorithm
    • 3.1.1.4 Support Vector Machine Algorithm
  • 3.1.2 Data Preparation for Classification
  • • 3.1.2.1 Outliers
    • 3.1.2.2 NULL Values
    • 3.1.2.3 Normalization
  • 3.1.3 Costs
  • 3.1.4 Priors
• 3.2 Regression
• • 3.2.1 Algorithm for Regression
• 3.3 Attribute Importance
• • 3.3.1 Data Preparation for Attribute Importance
  • 3.3.2 Algorithm for Attribute Importance
  • • 3.3.2.1 Minimum Description Length Algorithm
• 3.4 Anomaly Detection
• • 3.4.1 Algorithm for Anomaly Detection
  • • 3.4.1.1 Specify the One-Class SVM Algorithm
• 3.5 Testing Supervised Models
• • 3.5.1 Confusion Matrix
  • 3.5.2 Lift
  • 3.5.3 Receiver Operating Characteristics
  • 3.5.4 Test Metrics for Regression Models

4 Unsupervised Data Mining

• 4.1 Clustering
• • 4.1.1 Algorithms for Clustering
  • • 4.1.1.1 Enhanced k-Means Algorithm
    • 4.1.1.2 Orthogonal Partitioning Clustering (O-Cluster) Algorithm
    • 4.1.1.3 Outliers and Clustering
    • 4.1.1.4 K-Means and O-Cluster Comparison
• 4.2 Association
• • 4.2.1 Data for Association Models
  • 4.2.2 Difficult Cases for Associations
  • • 4.2.2.1 Finding Associations Involving Rare Events
  • 4.2.3 Algorithm for Associations
• 4.3 Feature Extraction
• • 4.3.1 Algorithm for Feature Extraction
  • • 4.3.1.1 NMF for Text Mining
    • 4.3.1.2 Data Preparation for NMF

5 Data Mining Process

• 5.1 How Is Data Mining Done?
• 5.2 How Does Oracle Data Mining Support Data Mining?
• • 5.2.1 Java and PL/SQL Interfaces
  • 5.2.2 Automated Data Mining
  • 5.2.3 Data Mining Functions
  • 5.2.4 Graphical Interfaces
  • 5.2.5 Model Deployment

6 Text Mining Using Oracle Data Mining

• 6.1 What is Text Mining?
• • 6.1.1 Document Classification
• 6.2 ODM Support for Text Mining
• • 6.2.1 Classification and Text Mining
  • 6.2.2 Clustering and Text Mining
  • 6.2.3 Feature Extraction and Text Mining
  • 6.2.4 Association and Text Mining
  • 6.2.5 Regression and Text Mining
  • 6.2.6 Anomaly Detection and Text Mining
• 6.3 Oracle Support for Text Mining

7 Oracle Data Mining Scoring Engine

• 7.1 ODM Scoring Engine Features
• 7.2 ODM Scoring Engine Installation
• 7.3 Scoring in Data Mining Applications
• 7.4 Moving Data Mining Models
• 7.5 Using the Oracle Data Mining Scoring Engine

8 Sequence Similarity Search and Alignment (BLAST)

• 8.1 Bioinformatics Sequence Search and Alignment
• 8.2 BLAST in the Oracle Database
• 8.3 Oracle Data Mining Sequence Search and Alignment Capabilities

Glossary

Index