IBM SPSS Modeler Essentials · Effective techniques for building powerful data mining and predictive analytics solutions by Salcedo, Jesus -- Read -- Imperial Library of Trantor

Index

Preface

What this book covers What you need for this book Who this book is for Conventions Reader feedback Customer support

Downloading the example code Downloading the color images of this book Errata Piracy Questions

Introduction to Data Mining and Predictive Analytics

Introduction to data mining CRISP-DM overview

Business Understanding Data Understanding Data Preparation Modeling Evaluation Deployment

Learning more about CRISP-DM

The data mining process (as a case study) Summary

The Basics of Using IBM SPSS Modeler

Introducing the Modeler graphic user interface

Stream canvas Palettes Modeler menus Toolbar Manager tabs Project window

Building streams

Mouse buttons Adding nodes Editing nodes Deleting nodes Building a stream Connecting nodes Deleting connections

Modeler stream rules Help options

Help menu Dialog help

Summary

Importing Data into Modeler

Data structure

Var. File source node Var. File source node File tab Var. File source node Data tab Var. File source node Filter tab Var. File source node Types tab Var. File source node Annotations tab Viewing data Excel source node Database source node

Levels of measurement and roles Summary

Data Quality and Exploration

Data Audit node options

Data Audit node results

The Quality tab

Missing data

Ways to address missing data Defining missing values in the Type node Imputing missing values with the Data Audit node

Summary

Cleaning and Selecting Data

Selecting cases

Expression Builder

Sorting cases Identifying and removing duplicate cases Reclassifying categorical values Summary

Combining Data Files

Combining data files with the Append node Removing fields with the Filter node Combining data files with the Merge node

The Filter tab The Optimization tab

Summary

Deriving New Fields

Derive – Formula Derive – Flag Derive – Nominal Derive – Conditional Summary

Looking for Relationships Between Fields

Relationships between categorical fields

Distribution node Matrix node

Relationships between categorical and continuous fields

Histogram node Means node

Relationships between continuous fields

Plot node Statistics node

Summary

Introduction to Modeling Options in IBM SPSS Modeler

Classification

Categorical targets Numeric targets The Auto nodes Data reduction modeling nodes

Association Segmentation

Choosing between models

Summary

Decision Tree Models

Decision tree theory CHAID theory

How CHAID processes different types of input variables Stopping rules Building a CHAID Model Partition node Overfitting CHAID dialog options

CHAID results

Summary

Model Assessment and Scoring

Contrasting model assessment with the Evaluation phase

Model assessment using the Analysis node Modifying CHAID settings Model comparison using the Analysis node Model assessment and comparison using the Evaluation node Scoring new data Exporting predictions

Summary

← Prev
Back
Next →

← Prev
Back
Next →