Data Mining: Concepts, Models and Techniques (Intelligent Systems Reference Library (12))

"The knowledge discovery process is as old as Homo sapiens. Until some time ago this process was solely based on the 'natural personal' computer provided by Mother Nature. Fortunately, in recent decades the problem has begun to be solved based on the development of the Data mining technology, aided by the huge computational power of the 'artificial' computers. Digging intelligently in different large databases, data mining aims to extract implicit, previously unknown and potentially useful information from data, since "knowledge is power". The goal of this book is to provide, in a friendly way, both theoretical concepts and, especially, practical techniques of this exciting field, ready to be applied in real-world situations. Accordingly, it is meant for all those who wish to learn how to explore and analysis of large quantities of data in order to discover the hidden nugget of information."--Back cover Subject Title 2 Preface 7 Contents 9 Introduction to Data Mining 12 What Is and What Is Not Data Mining? 12 Why Data Mining? 16 How to Mine the Data? 18 Problems Solvable with Data Mining 25 Classification 26 Cluster Analysis 30 Association Rule Discovery 34 Sequential Pattern Discovery 36 Regression 36 Deviation/Anomaly Detection 37 About Modeling and Models 37 Data Mining Applications 49 Data Mining Terminology 53 Privacy Issues 53 The “Data-Mine” 55 What Are Data? 55 Types of Datasets 56 Data Quality 60 Types of Attributes 62 Exploratory Data Analysis 67 What Is Exploratory Data Analysis? 67 Descriptive Statistics 69 Descriptive Statistics Parameters 70 Descriptive Statistics of a Couple of Series 78 Graphical Representation of a Dataset 91 Analysis of Correlation Matrix 95 Data Visualization 99 Examination of Distributions 109 Advanced Linear and Additive Models 115 Multiple Linear Regression 115 Logistic Regression 126 Cox Regression Model 130 Additive Models 133 Time Series: Forecasting 134 Multivariate Exploratory Techniques 140 Factor Analysis 140 Principal Components Analysis 143 Canonical Analysis 146 Discriminant Analysis 147 OLAP 148 Anomaly Detection 158 Classification and Decision Trees 168 What Is a Decision Tree? 168 Decision Tree Induction 170 GINI Index 175 Entropy 178 Misclassification Measure 180 Practical Issues Regarding Decision Trees 188 Predictive Accuracy 188 STOP Condition for Split 188 Pruning Decision Trees 189 Extracting Classification Rules from Decision Trees 191 Advantages of Decision Trees 192 Data Mining Techniques and Models 193 Data Mining Methods 193 Bayesian Classifier 194 Artificial Neural Networks 199 Perceptron 200 Types of Artificial Neural Networks 213 Probabilistic Neural Networks 225 Some Neural Networks Applications 232 Support Vector Machines 242 Association Rule Mining 257 Rule-Based Classification 260 k-Nearest Neighbor 264 Rough Sets 268 Clustering 279 Hierarchical Clustering 290 Non-hierarchical/Partitional Clustering 292 Genetic Algorithms 297 Components of GAs 300 Architecture of GAs 318 Applications 321 Classification Performance Evaluation 326 Costs and Classification Accuracy 326 ROC (Receiver Operating Characteristic) Curve 330 Statistical Methods for Comparing Classifiers 335 References 338 Index 360