Data Mining (eBook)

Yong Yin has been Associate Professor at Yamagata University, Japan, since 2004. He was previously Assistant Professor at the same university from 2002 to 2004. His research areas are manufacturing strategy; product development; workforce agility; and supply chain management.Ikou Kaku is a professor at the Department of Management Science and Engineering, Akita Prefectural University, Japan. His research interests are in human factors related to manufacturing; mathematical modeling and meta heuristics; data mining techniques and their application in inventory management; and supply chain management.Jiafu Tang is a professor at Northeastern University, Shenyang, China. He works in the Institute of Systems Engineering's Key Laboratory of Integrated Automation of Process Industry of MOE.JianMing Zhu is a professor at the Central University of Finance and Economics, Beijing, China. He works in the School of Information.

Preface 6
Contents 10
1 Decision Analysis and Cluster Analysis 16
1.1 Decision Tree 16
1.2 Cluster Analysis 19
References 23
2 Association Rules Mining in Inventory Database 24
2.1 Introduction 24
2.2 Basic Concepts of Association Rule 26
2.3 Mining Association Rules 29
2.3.1 The Apriori Algorithm: Searching Frequent Itemsets 29
2.3.2 Generating Association Rules from Frequent Itemsets 31
2.4 Related Studies on Mining Association Rulesin Inventory Database 32
2.4.1 Mining Multidimensional Association Rulesfrom Relational Databases 32
2.4.2 Mining Association Rules with Time-window 34
2.5 Summary 37
References 38
3 Fuzzy Modeling and Optimization: Theory and Methods 39
3.1 Introduction 39
3.2 Basic Terminology and Definition 41
3.2.1 Definition of Fuzzy Sets 41
3.2.2 Support and Cut Set 42
3.2.3 Convexity and Concavity 42
3.3 Operations and Properties for Generally Used Fuzzy Numbers 43
3.3.1 Fuzzy Inequality with Tolerance 43
3.3.2 Interval Numbers 44
3.3.3 L–R Type Fuzzy Number 45
3.3.4 Triangular Type Fuzzy Number 45
3.3.5 Trapezoidal Fuzzy Numbers 46
3.4 Fuzzy Modeling and Fuzzy Optimization 47
3.5 Classification of a Fuzzy Optimization Problem 49
3.5.1 Classification of the Fuzzy Extreme Problems 49
3.5.2 Classification of the Fuzzy Mathematical Programming Problems 50
3.5.3 Classification of the Fuzzy Linear Programming Problems 53
3.6 Brief Summary of Solution Methods for FOP 54
3.6.1 Symmetric Approaches Based on Fuzzy Decision 55
3.6.2 Symmetric Approach Based on Non-dominated Alternatives 57
3.6.3 Asymmetric Approaches 57
3.6.4 Possibility and Necessity Measure-based Approaches 60
3.6.5 Asymmetric Approaches to PMP5 and PMP6 61
3.6.6 Symmetric Approaches to the PMP7 63
3.6.7 Interactive Satisfying Solution Approach 63
3.6.8 Generalized Approach by Angelov 64
3.6.9 Fuzzy Genetic Algorithm 64
3.6.10 Genetic-based Fuzzy Optimal Solution Method 65
3.6.11 Penalty Function-based Approach 65
References 65
4 Genetic Algorithm-based Fuzzy Nonlinear Programming 69
4.1 GA-based Interactive Approach for QP Problemswith Fuzzy Objective and Resources 69
4.1.1 Introduction 69
4.1.2 Quadratic Programming Problems with Fuzzy Objective/Resource Constraints 70
4.1.3 Fuzzy Optimal Solution and Best Balance Degree 73
4.1.4 A Genetic Algorithm with Mutation Along the Weighted Gradient Direction 74
4.1.5 Human–Computer Interactive Procedure 76
4.1.6 A Numerical Illustration and Simulation Results 78
4.2 Nonlinear Programming Problems with Fuzzy Objectiveand Resources 80
4.2.1 Introduction 80
4.2.2 Formulation of NLP Problems with Fuzzy Objective/Resource Constraints 81
4.2.3 Inexact Approach Based on GA to Solve FO/RNP-1 84
4.2.4 Overall Procedure for FO/RNP by Meansof Human–Computer Interaction 86
4.2.5 Numerical Results and Analysis 88
4.3 A Non-symmetric Model for Fuzzy NLP Problemswith Penalty Coefficients 90
4.3.1 Introduction 90
4.3.2 Formulation of Fuzzy Nonlinear Programming Problems with Penalty Coefficients 90
4.3.3 Fuzzy Feasible Domain and Fuzzy Optimal Solution Set 93
4.3.4 Satisfying Solution and Crisp Optimal Solution 94
4.3.5 General Scheme to Implement the FNLP-PC Model 97
4.3.6 Numerical Illustration and Analysis 98
4.4 Concluding Remarks 99
References 100
5 Neural Network and Self-organizing Maps 101
5.1 Introduction 101
5.2 The Basic Concept of Self-organizing Map 103
5.3 The Trial Discussion on Convergence of SOM 106
5.4 Numerical Example 110
5.5 Conclusion 114
References 114
6 Privacy-preserving Data Mining 115
6.1 Introduction 115
6.2 Security, Privacy and Data Mining 118
6.2.1 Security 118
6.2.2 Privacy 119
6.2.3 Data Mining 121
6.3 Foundation of PPDM 123
6.3.1 The Characters of PPDM 123
6.3.2 Classification of PPDM Techniques 124
6.4 The Collusion Behaviors in PPDM 128
6.5 Summary 132
References 132
7 Supply Chain Design Using Decision Analysis 134
7.1 Introduction 134
7.2 Literature Review 136
7.3 The Model 137
7.4 Comparative Statics 140
7.5 Conclusion 144
References 144
8 Product Architecture and Product Development Processfor Global Performance 146
8.1 Introduction and Literature Review 146
8.2 The Research Problem 149
8.3 The Models 153
8.3.1 Two-function Products 153
8.3.2 Three-function Products 155
8.4 Comparisons and Implications 159
8.4.1 Three-function Products with Two Interfaces 159
8.4.2 Three-function Products with Three Interfaces 159
8.4.3 Implications 164
8.5 A Summary of the Model 165
8.6 Conclusion 167
References 167
9 Application of Cluster Analysis to Cellular Manufacturing 169
9.1 Introduction 169
9.2 Background 172
9.2.1 Machine-part Cell Formation 172
9.2.2 Similarity Coefficient Methods (SCM) 173
9.3 Why Present a Taxonomy on Similarity Coefficients? 173
9.3.1 Past Review Studies on SCM 174
9.3.2 Objective of this Study 174
9.3.3 Why SCM Are More Flexible 175
9.4 Taxonomy for Similarity Coefficients Employed in Cellular Manufacturing 177
9.5 Mapping SCM Studies onto the Taxonomy 181
9.6 General Discussion 188
9.6.1 Production Information-based Similarity Coefficients 188
9.6.2 Historical Evolution of Similarity Coefficients 191
9.7 Comparative Study of Similarity Coefficients 192
9.7.1 Objective 192
9.7.2 Previous Comparative Studies 193
9.8 Experimental Design 194
9.8.1 Tested Similarity Coefficients 194
9.8.2 Datasets 195
9.8.3 Clustering Procedure 199
9.8.4 Performance Measures 200
9.9 Comparison and Results 203
9.10 Conclusions 209
References 210
10 Manufacturing Cells Design by Cluster Analysis 218
10.1 Introduction 218
10.2 Background, Difficulty and Objective of this Study 220
10.2.1 Background 220
10.2.2 Objective of this Study and Drawbacksof Previous Research 222
10.3 Problem Formulation 224
10.3.1 Nomenclature 224
10.3.2 Generalized Similarity Coefficient 226
10.3.3 Definition of the New Similarity Coefficient 227
10.3.4 Illustrative Example 230
10.4 Solution Procedure 232
10.4.1 Stage 1 232
10.4.2 Stage 2 233
10.5 Comparative Study and Computational Performance 236
10.5.1 Problem 1 237
10.5.2 Problem 2 238
10.5.3 Problem 3 239
10.5.4 Computational Performance 240
10.6 Conclusions 240
References 241
11 Fuzzy Approach to Quality Function Deployment-based Product Planning 243
11.1 Introduction 243
11.2 QFD-based Integration Model for New Product Development 245
11.2.1 Relationship Between QFD Planning Process and Product Development Process 245
11.2.2 QFD-based Integrated Product Development ProcessModel 245
11.3 Problem Formulation of Product Planning 247
11.4 Actual Achieved Degree and Planned Degree 249
11.5 Formulation of Costs and Budget Constraint 249
11.6 Maximizing Overall Customer Satisfaction Model 251
11.7 Minimizing the Total Costs for Preferred Customer Satisfaction 253
11.8 Genetic Algorithm-based Interactive Approach 254
11.8.1 Formulation of Fuzzy Objective Function by Enterprise Satisfaction Level 254
11.8.2 Transforming FP2 into a Crisp Model 255
11.8.3 Genetic Algorithm-based Interactive Approach 256
11.9 Illustrated Example and Simulation Results 257
References 259
12 Decision Making with Consideration of Associationin Supply Chains 260
12.1 Introduction 260
12.2 Related Research 262
12.2.1 ABC Classification 262
12.2.2 Association Rule 262
12.2.3 Evaluating Index 263
12.3 Consideration and the Algorithm 264
12.3.1 Expected Dollar Usage of Item(s) 264
12.3.2 Further Analysis on EDU 265
12.3.3 New Algorithm of Inventory Classification 267
12.3.4 Enhanced Apriori Algorithm for Association Rules 267
12.3.5 Other Considerations of Correlation 269
12.4 Numerical Example and Discussion 270
12.5 Empirical Study 272
12.5.1 Datasets 272
12.5.2 Experimental Results 272
12.6 Concluding Remarks 276
References 276
13 Applying Self-organizing Maps to Master Data Makingin Automatic Exterior Inspection 278
13.1 Introduction 278
13.2 Applying SOM to Make Master Data 280
13.3 Experiments and Results 285
13.4 The Evaluative Criteria of the Learning Effect 286
13.4.1 Chi-squared Test 288
13.4.2 Square Measure of Close Loops 288
13.4.3 Distance Between Adjacent Neurons 289
13.4.4 Monotony of Close Loops 289
13.5 The Experimental Results of Comparing the Criteria 290
13.6 Conclusions 292
References 293
14 Application for Privacy-preserving Data Mining 294
14.1 Privacy-preserving Association Rule Mining 294
14.1.1 Privacy-preserving Association Rule Miningin Centralized Data 294
14.1.2 Privacy-preserving Association Rule Mining in Horizontal Partitioned Data 296
14.1.3 Privacy-preserving Association Rule Mining in Vertically Partitioned Data 297
14.2 Privacy-preserving Clustering 302
14.2.1 Privacy-preserving Clustering in Centralized Data 302
14.2.2 Privacy-preserving Clusteringin Horizontal Partitioned Data 302
14.2.3 Privacy-preserving Clustering in Vertically Partitioned Data 304
14.3 A Scheme to Privacy-preserving Collaborative Data Mining 307
14.3.1 Preliminaries 307
14.3.2 The Analysis of the Previous Protocol 309
14.3.3 A Scheme to Privacy-preserving Collaborative Data Mining 311
14.3.4 Protocol Analysis 312
14.4 Evaluation of Privacy Preservation 315
14.5 Conclusion 317
References 317
Index 319