Data Mining : Methods and Techniques

This text is written at an introductory level and provides real world problem analysis and practical examples as well as data mining theory to enable students to apply the concepts of data mining outside the classroom. This mix of theory and practical application makes this text directly suitable for a wide variety of undergraduate disciplines such as business computer science, students engineering, economics, statistics, marketing, information systems and information technology.

Table of Contents:
1. Fundamentals of data mining 1.1 Data mining: What is it? 1.1.1 Data query versus data mining 1.1.2 The learning process 1.2 The history of data mining 1.3 Data mining strategies 1.3.1 Classification 1.3.2 Association 1.3.3 Clustering 1.3.4 Estimation 1.3.5 Novelty detection 1.3.6 Sequence detection 1.4 Popular data mining techniques 1.4.1 Function estimation-based algorithms 1.4.2 Lazy learning-based algorithms 1.4.3 Meta learning-based algorithms 1.4.4 Probability-based algorithms 1.4.5 Tree-based algorithms 1.5 Data mining applications 1.6 Challenges of data mining 1.7 The future of data mining 1.8 Key terms 1.9 Further reading 1.10 Chapter references 1.11 Exercises 1.11.1 Review questions 1.11.2 Laboratory questions 2. Data processing and data warehouse 2.1 Data, information and knowledge: What are these? 2.2 Types of data 2.3 Data warehouse 2.4 Data cleaning 2.4.1 Missing data 2.4.2 Noisy data 2.5 Data de-normalization 2.6 Data transformation 2.7 Data quality measure 2.8 OLAP: What is it? 2.9 Data sampling 2.10 Key terms 2.11 Further reading 2.12 Chapter references 2.13 Exercises 2.13.1 Review questions 2.13.2 Laboratory questions 3. WEKA: An attractive data mining tool 3.1 Introduction 3.2 Installing WEKA 3.3 WEKA data file format 3.4 Starting WEKA 3.5 Data visualization 3.6 Data filtering 3.7 Selecting attributes 3.7.1 Factor analysis 3.8 Data mining with WEKA 3.9 Some other popular data mining packages 3.10 Key terms 3.11 Further reading 3.12 Chapter references 3.13 Exercises 3.13.1 Review questions 3.13.2 Laboratory questions 4. The classification task 4.1 Introduction 4.2 Decision Trees 4.2.1 Practical Considerations in Decision Trees 4.2.1.1 Avoiding an Over-Fitted Model 4.2.1.2 Categorizing Continuous Attributes 4.2.1.3 Dealing with Missing Attribute Values 4.2.2 Decision Tree Construction through WEKA 4.2.2.1 The Result List 4.2.2.2 Analysis of Results 4.2.3 Strengths and Weaknesses of Decision Trees 4.2.4 Applications of Decision Tree Algorithms 4.3 Probability-Based Solver: Naive Bayes 4.3.1 Application of Naive Bayes Algorithm in Weka 4.3.2 Strengths and Weaknesses of Naive Bayes Algorithm 4.3.3 Applications of Naive Bayes Algorithm 4.4 Key Terms 4.6 Further Reading 4.7 Chapter References 4.8 Exercises 4.8.1 Review Questions 4.8.2 Laboratory Questions 5. Artificial Neural Networks and Support Vector Machines 5.1 Artificial Neural Networks 5.1.1 Feed-Forward Neural Networks 5.1.1.1 Back Propagation Networks 5.1.2 Neural Network Construction Through WEKA 5.1.2.1 Parameter Value Selection for Neural Networks 5.1.3 Strengths and Weaknesses of Artificial Neural Networks 5.1.4 Applications of Neural Network Algorithms 5.2 Statistical Learning-Based Solver: Support Vector Machine 5.2.1 SVM Model Construction Through WEKA 5.2.2 Strengths and Weaknesses of SVM 5.2.3 Applications of SVMs 5.3 Key Terms 5.4 Further Reading 5.5 Chapter References 5.6 Exercises 5.6.1 Review Questions 5.6.2 Laboratory Questions 6. Model Evaluation Techniques 6.1 Introduction 6.2 Accuracy/Error Rate Estimation 6.2.1 The Confusion Matrix 6.3 Receiver Operating Curves 6.3.1 Area-based Accuracy Measure 6.3.2 Euclidean Distance Comparison 6.4 Lift Chart 6.5 Cost and Utility 6.6 Parsimony 6.7 Bagging and Boosting 6.7.1 Bagging and Boosting Performance Evaluation Through WEKA 6.8 The Model Ranking Approach 6.8.1 Model Ranking Through WEKA 6.9 Key Terms 6.10 Further Reading 6.11 Chapter References 6.12 Exercises 6.12.1 Review Questions 6.12.2 Laboratory Questions 7. Association Rule Mining 7.1 Transaction Data 7.2 Concepts of Association Rules 7.3 Relevance of Association Rule Mining 7.4 The functions of Association Rule Mining 7.4.1 Single-dimensional association rules 7.4.2 Multidimensional Association Rules 7.5 Two Other Parameters: Improvement and Share 7.6 Taxonomies 7.7 The problem of large datasets 7.8 Apriori Algorithm 7.9 Application of Apriori Algorithm through WEKA 7.10 Strengths and Weaknesses of ARM 7.11 Applications of ARM 7.12 Key Terms 7.13 Further Reading 7.14 Chapter References 7.15 Exercises 7.15.1 Review Questions 7.15.2 Laboratory Questions 8. The Clustering Task 8.1 Introduction 8.2 Distance Measures 8.2.1 Geometrical Distance Measures 8.2.2 Percent Disagreement 8.3 Types of Clusterin