Data Mining for Business Intelligence : Concepts, Techniques, and Applications in Microsoft Office Excel with XLMiner

by Shmueli, Galit; Patel, Nitin R.; Bruce, Peter C.

• ISBN13: 9780470526828
• ISBN10: 0470526823
• Edition: 2nd
• Format: Hardcover
• Copyright: 2010-10-26
• Publisher: Wiley

Summary

Data Mining for Business Intelligence, Second Edition uses real data and actual cases to illustrate the applicability of data mining (DM) intelligence in the development of successful business models. Featuring complimentary access to XLMiner®, the Microsoft Office Excel® add-in, this book allows readers to follow along and implement algorithms at their own speed, with a minimal learning curve. In addition, students and practitioners of DM techniques are presented with hands-on, business-oriented applications. An abundant amount of exercises and examples, now doubled in number in the second edition, are provided to motivate learning and understanding.

This book helps readers understand the beneficial relationship that can be established between DM and smart business practices, and is an excellent learning tool for creating valuable strategies and making wiser business decisions. New topics include detailed coverage of visualization enhanced by Spotfire subroutines and time series forecasting, among a host of other subject matter.

The Second Edition now features:

-Three new chapters on time series forecasting, introducing popular business forecasting methods including moving average, exponential smoothing methods; regression-based models; and topics such as explanatory vs. predictive modeling, two-level models, and ensembles

-A revised chapter on data visualization that now features interactive visualization principles and added assignments that demonstrate interactive visualization in practice

-Separate chapters that each treat k-nearest neighbors and Naïve Bayes methods

-Summaries at the start of each chapter that supply an outline of key topics

Author Biography

GALIT SHMUELI, PhD, is Associate Professor of Statistics and Director of the eMarkets Research Lab in the Robert H. Smith School of Business at the University of Maryland. Dr. Shmueli is the coauthor of Statistical Methods in e-Commerce Research and Modeling Online Auctions, both published by Wiley.

NITIN R. PATEL, PhD, is Chairman and cofounder of Cytel, Inc., based in Cambridge, Massachusetts. A Fellow of the American Statistical Association, Dr. Patel has also served as a Visiting Professor at the Massachusetts Institute of Technology for over ten years.

PETER C. BRUCE is President and owner of statistics.com, the leading provider of online education in statistics.

Table of Contents

Foreword

Preface

Acknowledgments

Preliminaries

Introduction

What Is Data Mining?

Where Is Data Mining Used?

The Origins of Data Mining

The Rapid Growth of Data Mining

Why Are There So Many Different Methods?

Terminology and Notation

Road Maps to This Book

Overview of the Data Mining Process

Introduction

Core Ideas in Data Mining

Supervised and Unsupervised Learning

The Steps in Data Mining

Preliminary Steps

Building a Model: Example with Linear Regression

Using Excel for Data Mining

Problems

Data Exploration and Dimension Reduction

Data Visualization

Uses of Data Visualization

Data Examples

Boston Housing Data

Ridership on Amtrak Trains

Basic Charts: bar charts, line graphs, and scatterplots

Distribution Plots

Heatmaps: visualizing correlations and missing values

MultiDimensional Visualization

Adding Variables: color, hue, size, shape, multiple panels, animation

Manipulations: rescaling,aggregation and hierarchies, zooming and panning, filtering

Reference: trend line and labels

Scaling up: large datasets

Multivariate plot: parallel coordinates plot

Interactive visualization

Specialized Visualizations

Visualizing networked data

Visualizing hierarchical data: treemaps

Visualizing geographical data: maps

Summary of major visualizations and operations, according to data mining goal

Prediction

Classification

Time series forecasting

Unsupervised learning

Problems

Dimension Reduction

Introduction

Practical Considerations

House Prices in Boston

Data Summaries

Correlation Analysis

Reducing the Number of Categories in Categorical Variables

Converting A Categorical Variable to A Numerical Variable

Principal Components Analysis

Breakfast Cereals

Principal Components

Normalizing the Data

Using Principal Components for Classification and Prediction

Dimension Reduction Using Regression Models

Dimension Reduction Using Classification and Regression Trees

Problems

Performance Evaluation

Evaluating Classification and Predictive Performance

Introduction

Judging Classification Performance

Benchmark: The Naive Rule

Class Separation

The Classification Matrix

Using the Validation Data

Accuracy Measures

Cutoff for Classification

Performance in Unequal Importance of Classes

Asymmetric Misclassification Costs

Oversampling and Asymmetric Costs

Classification Using a Triage Strategy

Evaluating Predictive Performance

Benchmark: The Average

Prediction Accuracy Measures

Problems

Prediction and Classification Methods

Multiple Linear Regression

Introduction

Explanatory vs. Predictive Modeling

Estimating the Regression Equation and Prediction

Example: Predicting the Price of Used Toyota Corolla Automobiles

Variable Selection in Linear Regression

Reducing the Number of Predictors

How to Reduce the Number of Predictors

Problems

kNearest

Neighbors (kNN)

The kNN

Classifier

Determining Neighbors

Classification Rule

Example: Riding Mowers

Choosing k

Setting the Cutoff Value

kNN

With More Than 2 Classes

kNN

for a Numerical Response

Advantages and Shortcomings of kNN

Algorithms

Problems

Naive Bayes

Introduction

Predicting Fraudulent Financial Reporting

The Practical Difficulty with the Complete (Exact) Bayes Procedure

The Solution: Na‹ve Bayes

Predicting Fraudulent Financial Reports, 2 Predictors

Predicting Delayed Flights

Advantages and Shortcomings of the naive Bayes Classifier

Problems

Classification and Regression Trees

Introduction

Classification Trees

Recursive Partitioning

Riding Mowers

Measures of Impurity

Evaluating the Performance of a Classification Tree

Acceptance of Personal Loan

Avoiding Overfitting

Stopping Tree Growth: CHAID

Pruning the Tree

Classification Rules from Trees

Classification Trees for More Than 2 Classes

Regression Trees

Prediction

Measuring Impurity

Evaluating Performance

Advantages, Weaknesses, and Extensions

Problems

Logistic Regression

Introduction

The Logistic Regression Model

Example: Acceptance of Personal Loan

Model with a Single Predictor

Estimating the Logistic Model from Data: Computing Parameter

Estimates

Interpreting Results in Terms of Odds

Evaluating Classification Performance

Variable Selection

Example of Complete Analysis: Predicting Delayed Flights

Data Preprocessing

Model Fitting and Estimation

Model Interpretation

Model Performance

Variable Selection

Appendix: Logistic Regression for Profiling

Appendix: Logistic regression for profiling

Appendix: B: Evaluating Goodness of Fit

Appendix B Evaluating Goodness of Fit

Appendix: C: Logistic Regression for More Than Two Classes

Appendix C Logistic Regression for More Than Two Classes

Problems

Neural Nets

Introduction

Concept and Structure of a Neural Network

Fitting a Network to Data

Tiny Dataset

Computing Output of Nodes

Preprocessing the Data

Training the Model

Classifying Accident Severity

Avoiding overfitting

Using the Output for Prediction and Classification

Required User Input

Exploring the Relationship Between Predictors and Response

Advantages and Weaknesses of Neural Networks

Problems

Discriminant Analysis

Introduction

Riding Mowers

Personal Loan Acceptance

Distance of an Observation from a Class

Fisher's Linear Classification Functions

Classification Performance of Discriminant Analysis

Prior Probabilities

Unequal Misclassification Costs

Classifying More Than Two Classes

Medical Dispatch to Accident Scenes

Advantages and Weaknesses

Problems

Mining Relationships Among Records

Association Rules

Introduction

Discovering Association Rules in Transaction Databases

Synthetic Data on Purchases of Phone Faceplates

Generating Candidate Rules

The Apriori Algorithm

Selecting Strong Rules

Support and Confidence

Lift Ratio

Data Format

The Process of Rule Selection

Interpreting the Results

Statistical Significance of Rules

Rules for Similar Book Purchases

Summary

Problems

Cluster Analysis

Introduction

Example: Public Utilities

Measuring Distance Between Two Records

Euclidean Distance

Normalizing Numerical Measurements

Other Distance Measures for Numerical Data

Distance Measures for Categorical Data

Distance Measures for Mixed Data

Measuring Distance Between Two Clusters

Hierarchical (Agglomerative) Clustering

Contents

Minimum Distance (Single Linkage)

Maximum Distance (Complete Linkage)

Average Distance (Average Linkage)

Dendrograms: Displaying Clustering Process and Results

Validating Clusters

Limitations of Hierarchical Clustering

Nonhierarchical Clustering: The kMeans Algorithm

Initial Partition into k Clusters

Problems

Forecasting Time Series

Handling Time Series

Introduction

Explanatory vs. Predictive Modeling

Popular Forecasting Methods in Business

Combining Methods

Time Series Components

Example: Ridership on Amtrak Trains

Data Partitioning

Problems

Regression Based Forecasting

A Model with Trend

Linear Trend

Exponential Trend

Polynomial Trend

A Model with Seasonality

A model with trend and seasonality

Autocorrelation and ARIMA Models

Computing Autocorrelation

Computing Autocorrelation

Improving Forecasts by Integrating Autocorrelation Information

Improving Forecasts by Integrating Autocorrelation Information

Evaluating Predictability

Evaluating Predictability

Problems

Smoothing Methods

Introduction

Moving Average

Centered Moving Average for Visualization

Trailing Moving Average for Forecasting

Choosing Window Width

Simple Exponential Smoothing

Choosing Smoothing Parameter

Relation Between Moving Average and Simple Exponential

Smoothing

Advanced Exponential Smoothing

Series with a trend

Series with a trend and seasonality

Series with seasonality

Problems

Cases

Cases

Charles Book Club

German Credit

Tayko Software Cataloger

Segmenting Consumers of Bath Soap

DirectMail Fundraising

Catalog CrossSelling

Predicting Bankruptcy

Time Series Case: Forecasting Public Transportation Demand

References

Index

Table of Contents provided by Publisher. All Rights Reserved.