Contents

List of Examples

List of Figures

List of Tables

Title and Copyright Information

Preface

• Audience
• Documentation Accessibility
• Related Documentation
• Conventions

What's New in Oracle Data Mining?

• New Features in Oracle Data Mining 11g Release 2 (11.2)
• New Features in Oracle Data Mining 11g Release 1 (11.1)
• New Features in Oracle Data Mining 10g Release 2 (10.2)

Part I Introductions

1 What Is Data Mining?

• What Is Data Mining?
  • Automatic Discovery
  • Prediction
  • Grouping
  • Actionable Information
  • Data Mining and Statistics
  • Data Mining and OLAP
  • Data Mining and Data Warehousing
• What Can Data Mining Do and Not Do?
  • Asking the Right Questions
  • Understanding Your Data
• The Data Mining Process
  • Problem Definition
  • Data Gathering and Preparation
  • Model Building and Evaluation
  • Knowledge Deployment

2 Introducing Oracle Data Mining

• Data Mining in the Database Kernel
• Data Mining Functions
  • Supervised Data Mining
    • Supervised Learning: Testing
    • Supervised Learning: Scoring
  • Unsupervised Data Mining
    • Unsupervised Learning: Scoring
  • Oracle Data Mining Functions
• Data Mining Algorithms
  • Oracle Data Mining Supervised Algorithms
  • Oracle Data Mining Unsupervised Algorithms
• Data Preparation
  • Supermodels
• How Do I Use Oracle Data Mining?
  • Oracle Data Miner
  • PL/SQL Packages
  • SQL Functions
  • Java API
  • Oracle Spreadsheet Add-In for Predictive Analytics
• Where Do I Find Information About Oracle Data Mining?
  • Oracle Data Mining Resources on the Oracle Technology Network
  • Oracle Data Mining Publications
• Oracle Data Mining and Oracle Database Analytics

3 Introducing Oracle Predictive Analytics

• About Predictive Analytics
  • Predictive Analytics and Data Mining
  • How Does it Work?
  • Predictive Analytics Operations
• Oracle Spreadsheet Add-In for Predictive Analytics
• APIs for Predictive Analytics
  • Predictive Analytics in the PL/SQL API
  • Predictive Analytics in the Java API
    • Example: Use OraProfileTask to Create Profile Results
• Example: PREDICT
• Behind the Scenes
  • EXPLAIN
  • PREDICT
    • Accuracy
  • PROFILE

Part II Mining Functions

4 Regression

• About Regression
  • How Does Regression Work?
    • Linear Regression
    • Multivariate Linear Regression
    • Regression Coefficients
    • Nonlinear Regression
    • Multivariate Nonlinear Regression
    • Confidence Bounds
• A Sample Regression Problem
• Testing a Regression Model
  • Residual Plot
  • Regression Statistics
    • Root Mean Squared Error
    • Mean Absolute Error
    • Test Metrics in Oracle Data Miner
• Regression Algorithms

5 Classification

• About Classification
• A Sample Classification Problem
• Testing a Classification Model
  • Accuracy
  • Confusion Matrix
  • Lift
    • Lift Statistics
  • Receiver Operating Characteristic (ROC)
    • The ROC Curve
    • Area Under the Curve
    • ROC and Model Bias
    • ROC Statistics
• Biasing a Classification Model
  • Costs
    • Costs Versus Accuracy
    • Positive and Negative Classes
    • Assigning Costs and Benefits
  • Priors
• Classification Algorithms

6 Anomaly Detection

• About Anomaly Detection
  • One-Class Classification
  • Anomaly Detection for Single-Class Data
  • Anomaly Detection for Finding Outliers
• Sample Anomaly Detection Problems
  • Example: Find Outliers
  • Example: Score New Data
• Algorithm for Anomaly Detection

7 Clustering

• About Clustering
  • Interpreting Clusters
  • How are Clusters Computed?
  • Cluster Rules
    • Support and Confidence
    • Number of Clusters
  • Attribute Histograms
  • Centroid of a Cluster
  • Scoring New Data
• Sample Clustering Problems
  • Example: Find Clusters
  • Example: Score New Data
• Clustering Algorithms

8 Association

• About Association
  • Transactions
    • Items and Collections
    • Sparse Data
  • Itemsets
  • Frequent Itemsets
• A Sample Association Problem
• Algorithm for Association Rules

9 Feature Selection and Extraction

• Finding the Best Attributes
• Feature Selection
  • Data for Attribute Importance
  • Example: Attribute Importance
  • Example: Predictive Analytics EXPLAIN
• Feature Extraction
  • Data for Feature Extraction
  • Example: Features Created from Build Data
  • Example: Scored Features
• Algorithms for Feature Selection and Extraction

Part III Algorithms

10 Apriori

• About Apriori
  • Association Rules
    • Antecedent and Consequent
    • Confidence
    • Example: Calculating Rules from Frequent Itemsets
• Metrics for Association Rules
  • Support
  • Confidence
  • Lift
• Data for Association Rules

11 Decision Tree

• About Decision Tree
  • Decision Tree Rules
    • Confidence and Support
  • Advantages of Decision Trees
  • Growing a Decision Tree
    • Splitting
    • Cost Matrix
    • Preventing Over-Fitting
  • XML for Decision Tree Models
• Tuning the Decision Tree Algorithm
• Data Preparation for Decision Tree

12 Generalized Linear Models

• About Generalized Linear Models
  • GLM in Oracle Data Mining
    • Interpretability and Transparency
    • Wide Data
    • Confidence Bounds
  • Ridge Regression
    • Build Settings for Ridge Regression
    • Ridge and Confidence Bounds
    • Ridge and Variance Inflation Factor for Linear Regression
    • Ridge and Data Preparation
• Tuning and Diagnostics for GLM
  • Build Settings
  • Diagnostics
    • Coefficient Statistics
    • Global Model Statistics
    • Row Diagnostics
• Data Preparation for GLM
  • Data Preparation for Linear Regression
  • Data Preparation for Logistic Regression
  • Missing Values
• Linear Regression
  • Coefficient Statistics for Linear Regression
  • Global Model Statistics for Linear Regression
  • Row Diagnostics for Linear Regression
• Logistic Regression
  • Reference Class
  • Class Weights
  • Coefficient Statistics for Logistic Regression
  • Global Model Statistics for Logistic Regression
  • Row Diagnostics for Logistic Regression

13 k-Means

• About k-Means
• Data Preparation for k-Means

14 Minimum Description Length

• About MDL
• Data Preparation for MDL

15 Naive Bayes

• About Naive Bayes
  • Advantages of Naive Bayes
• Tuning a Naive Bayes Model
• Data Preparation for Naive Bayes

16 Non-Negative Matrix Factorization

• About NMF
  • How Does it Work?
• Data Preparation for NMF

17 O-Cluster

• About O-Cluster
• Data Preparation for O-Cluster
  • User-Specified Data Preparation for O-Cluster

18 Support Vector Machines

• About Support Vector Machines
  • Advantages of SVM
  • Advantages of SVM in Oracle Data Mining
    • Usability
    • Scalability
  • Kernel-Based Learning
  • Active Learning
• Tuning an SVM Model
• Data Preparation for SVM
  • Normalization
  • SVM and Automatic Data Preparation
• SVM Classification
  • Class Weights
• One-Class SVM
• SVM Regression

Part IV Data Preparation

19 Automatic and Embedded Data Preparation

• Overview
  • The Case Table
  • Data Type Conversion
    • Date Data
  • Text Transformation
  • Business and Domain-Sensitive Transformations
• Automatic Data Preparation
  • Enabling Automatic Data Preparation
  • Overview of Algorithm-Specific Transformations
    • Binning
    • Normalization
    • Outlier Treatment
  • Algorithms and ADP
• Embedded Data Preparation
  • Transformation Lists and ADP
  • Creating a Transformation List
    • Transforming a Nested Attribute
  • Oracle Data Mining Transformation Routines
    • Binning
    • Normalization
    • Outlier Treatment
• Transparency
  • Model Details and the Build Data
  • Reverse Transformations
    • Altering the Reverse Transformation Expression

Part V Mining Unstructured Data

20 Text Mining

• About Unstructured Data
• How Oracle Data Mining Supports Unstructured Data
  • Mixed Data
  • Text Data Types
  • Text Mining Algorithms
  • Text Classification
    • Multi-Class Document Classification
    • Multi-Target Document Classification
    • Document Classification Algorithms
  • Text Clustering
  • Text Feature Extraction
  • Text Association
  • Text Attribute Importance
• Preparing Text for Mining
• Sample Text Mining Problem
• Oracle Data Mining and Oracle Text

Glossary

Index