Computational Intelligence for Decision Support (International Series on Computational Intelligence)

Part IV......Page 0
Computational Intelligence for Decision Support......Page 1
Table of Contents......Page 6
WHAT READERS CAN EXPECT FROM THIS BOOK......Page 16
HOW THIS BOOK IS ORGANIZED......Page 17
For instructors:......Page 18
For scientists and leisure readers:......Page 19
1.2 THE NEED FOR DECISION SUPPORT AGENTS......Page 21
1.3 COMPUTERIZED DECISION SUPPORT MECHANISMS......Page 22
1.5 A REMARK ON TERMINOLOGY......Page 23
1.6 DATA,INFORMATION AND KNOWLEDGE......Page 25
1.7 ISSUES TO BE DISCUSSED IN THIS BOOK......Page 26
SELF-EXAMINATION QUESTIONS......Page 28
REFERENCES......Page 29
2.2.1 SOME SIMPLE EXAMPLES......Page 31
2.2.2 APPLICATIONS......Page 34
2.3.2 COMPUTATIONAL INTELLIGENCE AS AGENT-BASED PROBLEM SOVLING......Page 36
2.3.3 MEASURING THE INTELLIGENCE:TURING TEST......Page 37
2.4.1.2 Physically grounded......Page 38
2.4.2 SEQUENTIAL OR PARALLEL......Page 39
2.4.4 HUMAN INTELLIGENCE AS METAPHOR......Page 40
2.5.1 ABSTRACT DATA TYPES AND DATA STRUCTURES......Page 41
2.5.3 TREES......Page 42
2.5.6 GRAPHS......Page 43
2.6.1 MEANINGS OF SEARCH......Page 44
2.6.2 STATE SPACE SEARCH......Page 45
2.7.1 LEVELS OF ABSTRACTION IN COMPUTATIONAL INTELLIGENCE PROBLEM SOLVING......Page 46
2.7.2 USING ABSTRACT LEVELS......Page 47
2.7.3.2 Remarks on LISP,and Prolog and C++......Page 48
2.8.1.1 Depth-first search......Page 49
2.8.1.3 Iterative deepening search......Page 50
2.8.1.4 Comparison of uninformed search algorithms......Page 51
2.8.2.1 Heuristics......Page 52
2.8.2.2 Best first search......Page 53
2.9 REMARK ON CONSTRAINT-BASED SEARCH......Page 56
2.10.1 PLANNING AS SEARCH......Page 57
2.10.2 SYMBOL-BASED MACHINE LEARNING AS SEARCH......Page 58
SUMMARY......Page 59
SELF-EXAMINATION QUESTIONS......Page 60
REFERENCES......Page 61
3.2.1 BASICS......Page 63
3.2.2 PROPOSITIONAL CALCULUS......Page 64
3.2.3 PREDICATES......Page 65
3.2.5 KNOWLEDGE BASE......Page 67
3.2.6 INFERENCE RULES......Page 68
3.2.8 RESOLUTION --THE BASIC IDEA......Page 69
3.3.1.2 Structure of a Prolog statement......Page 73
3.3.1.3 Remarks on structure of a Prolog program......Page 74
3.3.1.4 Two kinds of queries (retrieval and confirmation)......Page 75
3.3.1.6 Answering query through depth first search......Page 76
3.3.1.8 Unification through recursion......Page 77
3.3.2.1 "I am my own grandfather"puzzle......Page 78
3.3.2.2 Farmer,wolf,goat and cabbage puzzle revisited......Page 79
3.3.3 SUMMARY OF IMPORTANT THINGS ABOUT PROLOG......Page 82
3.4.2 INDUCTION......Page 83
3.5.1 MEANING OF NONMONOTONIC REASONING......Page 84
3.5.2 COMMONSENSE REASONING......Page 85
3.5.3 CIRCUMSCRIPTION......Page 86
SUMMARY......Page 87
REFERENCES......Page 88
4.2 THE CONCEPT OF RELATION......Page 89
4.3.1 SCHEMA AND INSTANCE......Page 90
4.3.2 DECLARATIVE AND PROCEDURAL LANGUAGES......Page 91
4.4.1 PREVIEW OF RELATIONAL ALGEBRA......Page 92
4.4.2 HOW TO FORM A RELATIONAL ALGEBRA QUERY FROM A GIVEN ENGLISH QUERY......Page 93
4.4.4 RELATIONAL ALGEBRA:ADDITIONAL OPERATORS......Page 94
4.4.5 COMBINED USE OF OPERATORS......Page 96
4.5.1 VIRTUAL VIEWS AND MATERIALIZED VIEWS......Page 97
4.5.2 INTEGRITY CONSTRAINTS......Page 98
4.6 FUNCTIONAL DEPENDENCIES......Page 99
4.6.2 KEYS AND FUNCTIONAL DEPENDENCIES......Page 100
4.6.4 CLOSURES AND CANONICAL COVER......Page 101
4.6.5 ALGORITHMS FOR FINDING KEYS FROM FUNCTIONAL DEPENDENCIES......Page 102
4.7.1 WHAT IS THE MEANING OF A GOOD DESIGN AND WHY STUDY IT?......Page 103
4.7.2 BOYCE-CODD NORMAL FORM (BCNF)AND THIRD NORMAL FORM (3NF)......Page 105
4.7.3 REMARKS ON NORMAL FORMS AND DENORMALIZATION......Page 106
4.7.4.1 Lossless-join decomposition......Page 107
4.7.5 DECOMPOSITION ALGORITHMS......Page 108
4.8.1 VARIOUS FORMS OF DEPENDENCIES......Page 110
4.8.2.2 Important properties......Page 111
4.8.3.1 Definition of 4NF......Page 112
4.9 REMARK ON OBJECT-ORIENTED LOGICAL DATA MODELING......Page 113
4.10.2.1 EDB and IDB......Page 114
4.10.2.3 Recursive queries with negation in rule body:Using stratification......Page 115
4.10.3.1 Bottom-up versus top-down......Page 117
4.10.3.2 Magic sets approach for recursive query processing......Page 118
4.11 KNOWLEDGE REPRESENTATION MEETS DATABASES......Page 119
SUMMARY......Page 120
REFERENCES......Page 121
5.1 OVERVIEW......Page 123
5.2.2 THREE LEVELS OF DATA ABSTRACTION......Page 124
5.2.4 DATA MODELS......Page 125
5.3.1 BASIC REMARKS ON COMMERCIAL LANGUAGES......Page 126
5.3.3 EXAMPLES OF SQL QUERIES......Page 127
5.3.4 WRITING SIMPLE SQL QUERIES......Page 128
5.3.6 REMARKS ON INTEGRITY CONSTRAINTS......Page 129
5.3.8 REMARKS ON ENHANCEMENT OF SQL......Page 130
5.4.1 STORAGE MEDIA......Page 131
5.4.2 FILE STRUCTURES AND INDEXING......Page 132
5.5.1 QUERY PROCESSING......Page 133
5.5.3 HOW TRANSACTION PROCESSING IS RELATED TO QUERY PROCESSING......Page 134
5.6.2 BASICS OF INFORMATION RETRIEVAL......Page 135
5.6.3 WEB SEARCHING,DATABASE RETRIEVAL,AND IR......Page 137
5.7.1.1 Basics of parallel databases......Page 138
5.7.1.2 Distributed database systems......Page 139
5.7.2 DATA WAREHOUSING AND DECISION SUPPORT......Page 140
5.7.3 MIDDLEWARE......Page 141
5.8.1 FROM DATA AND INFORMATION RETRIEVAL TO KNOWLEDGE RETRIEVAL......Page 142
5.8.2 DEDUCTIVE RETRIEVAL SYSTEMS......Page 143
5.8.5 PRODUCTION SYSTEM MODEL......Page 144
5.8.5.2 The recognize-act cycle......Page 145
5.8.5.3 The need for a separate knowledge base......Page 147
5.8.6 KNOWLEDGE ENGINEERING......Page 148
5.8.7.2 Some important features of rule-based systems......Page 149
5.8.7.3 A simple example......Page 150
5.8.7.5 Explanation facility......Page 151
5.8.8.1 Weak methods,Strong methods and Role-limiting methods......Page 152
5.8.9 CLIPS:A BRIEF OVERVIEW......Page 153
5.9.1 WHAT IS KNOWLEDGE MANAGEMENT?......Page 154
5.9.2 INFORMATION TECHNOLGOY FOR KNOWLEDGE MANAGEMENT......Page 155
5.9.3 DATA AND KNOWLEDGE MANAGEMENT ONTOLOGIES......Page 156
SELF-EXAMINATION QUESTIONS......Page 157
REFERENCES......Page 158
6.2.1 WHAT IS THE ENTITY-RELATIONSHIP (ER)APPROACH?......Page 160
6.2.2 A SIMPLE EXAMPLE......Page 161
6.2.4 SOME IMPORTANT CONCEPTS......Page 162
6.2.5 DESIGN ISSUES IN ER MODELING......Page 163
6.2.7 KEYS IN CONVERTED TABLES......Page 164
6.2.8.3 Converting to tables......Page 165
6.2.9 EXTENDED ER FEATURES AND RELATIONSHIP WITH OBJECT-ORIENTED MODELING......Page 166
6.3 REMARK ON LEGACY DATA MODELS......Page 167
6.4 KNOWLEDGE MODELING FOR KNOWLEDGE REPRESENTATION......Page 168
6.5.1 SOME IMPORTANT ISSUES INVOLVED IN KNOWLEDGE REPRESENTATION AND REASONING......Page 169
6.6.1 BASICS OF FRAMES......Page 170
6.6.3.1 Inheritance in frame systems......Page 171
6.7.1 WHAT IS A CONCEPTUAL GRAPH?......Page 172
6.7.3 OPERATIONS......Page 174
6.7.4.1 Propositional node......Page 175
6.7.4.2 Inference rules......Page 176
6.7.4.3 Converting to predicate logic......Page 177
6.8 USER MODELING AND FLEXIBLE INFERENCE CONTROL......Page 178
SUMMARY......Page 179
REFERENCES......Page 180
7.2.1 SOME FORMS OF NON-EXACT RETRIEVAL......Page 181
7.2.2 BASICS OF ANALOGICAL REASONING......Page 183
7.3.2 STRUCTURE MAPPING FOR SUGGESTION-GENERATION......Page 184
7.3.3.1 Conversion of documents into unstructured databases......Page 185
7.3.3.2 Document algebra:an algebra on document stems and relations......Page 189
7.3.4.1 Basic idea and an example......Page 195
7.3.4.2 Steps for analogical problem solving......Page 197
7.3.4.3 Structure mapping for generating suggestions......Page 198
SELF-EXAMINATION QUESTIONS......Page 202
REFERENCES......Page 203
8.2.1 REMARKS ON CREATIVITY......Page 204
8.2.2 THEORETICAL FOUNDATION FOR STIMULATING HUMAN THINKING......Page 205
8.2.3 CREATIVITY IN DECISION SUPPORT SYSTEMS......Page 206
8.3.1 BASICS OF IDEA PROCESSORS......Page 207
8.3.3 HOW IDEA PROCESSORS WORK......Page 209
8.3.4 THE NATURE OF IDEA PROCESSORS......Page 210
8.4.1 RETROSPECTIVE ANALYSIS OF TECHNICAL INVENTION......Page 212
8.4.2 RETROSPECTIVE ANALYSIS FOR KNOWLEDGE-BASED IDEA GENERATION OF NEW ARTIFACTS......Page 214
8.4.3.1 Frames and inheritance in artifact representation......Page 215
8.5.2 STRATEGIC KNOWLEDGE AS KNOWLEDGE RELATED TO CREATIVITY......Page 218
8.5.3 STUDYING STRATEGIC HEURISTICS OF CREATIVE KNOWLEDGE......Page 220
8.5.4 DIFFICULTIES AND PROBLEMS IN ACQUIRING STRATEGIC HEURISTICS......Page 221
8.5.5 THE NATURE OF STRATEGIC HEURISTICS......Page 222
8.5.6 TOWARD KNOWLEDGE-BASED ARCHITECTURE COMBINING CREATIVITY AND EXPERTISE......Page 223
SUMMARY......Page 224
REFERENCES......Page 225
9.2.1 WHAT IS A QUESTION ANSWERING SYSTEM?......Page 228
9.3 INTENSIONAL ANSWERING AND CONCEPTUAL QUERY......Page 229
9.3.2 INTENSIONAL ANSWERING USING KNOWLEDGE DISCOVERY......Page 230
9.3.3 CONCEPTUAL QUERY ANSWERING......Page 232
9.3.4.1 The duality principle......Page 233
9.3.4.3 Query-invoked generation of intensional answers......Page 234
9.4.1 INTRODUCTION......Page 235
9.4.2 CONSTRUCTING AN ABSTRACT DATABASE FOR INTENSIONAL ANSWERS......Page 236
9.4.3 GENERATING INTENSIONAL ANSWERS FOR CONCEPTUAL QUERIES......Page 238
9.4.4 METHOD FOR INTENSIONAL CONCEPTUAL QUERY ANSWERING......Page 239
REFERENCES......Page 240
10.1 OVERVIEW......Page 242
10.2.1 MACHINE LEARNING:DEFINITION AND APPROACHES......Page 243
10.3.2 CANDIDATE ELIMINATION ALGORITHM......Page 244
10.3.3 ID3 ALGORITHM AND C4.5......Page 245
10.4.2.1 Why theory of learning is important......Page 249
10.5.1.1 Review of neural networks......Page 250
10.5.1.3 Unsupervised learning......Page 251
10.5.2.2 Genetic algorithms......Page 252
10.6.1 THE POPULARITY OF DATA MINING......Page 256
10.6.2 KDD VERSUS DATA MINING......Page 257
10.6.3 DATA MINING VERSUS MACHINE LEARNING......Page 259
10.6.4 DATA MINING VERSUS EXTENDED RETRIEVAL......Page 260
10.6.5 DATA MINING VERSUS STATISTIC ANALYSIS AND INTELLIGENT DATA ANALYSIS......Page 261
10.6.7 SUMMARY OF FEATURES......Page 262
10.7.2 DISCOVERY OR PREDICTION......Page 263
10.7.4 CLASSIFYING DATA MINING METHODS......Page 264
10.8.1 TERMINOLOGY......Page 265
10.8.2 FINDING ASSOCIATION RULES USING APRIORI ALGORITHM......Page 268
10.8.3.2 Sampling techniques in finding association rules......Page 270
10.8.3.4 Clustering and representative association rules......Page 271
SELF-EXAMINATION QUESTIONS......Page 272
REFERENCES......Page 273
11.1 OVERVIEW......Page 277
11.2 DATA MINING IN DATA WAREHOUSES......Page 278
11.3.1 DECISION SUPPORT QUERIES......Page 279
11.3.2.1 Components in data warehouses......Page 280
11.3.2.2 Relationship between data warehousing and OLAP......Page 281
11.3.3.1 Terminology......Page 282
11.3.3.2 OLAP operations......Page 283
11.3.3.4 Star schema and snowflake schema......Page 285
11.4.1 REVIEW OF A POPULAR DEFINITION......Page 286
11.4.2.1 The necessity of using materialized views......Page 287
11.4.2.2 The many facets of materialized views......Page 288
11.4.3 MAINTENANCE OF MATERIALIZED VIEWS......Page 289
11.4.4.1 Normalization versus denormalization......Page 290
11.4.5 INDEXING TECHNIQUES FOR IMPLEMENTATION......Page 291
11.5 REMARKS ON PHYSICAL DESIGN OF DATA WAREHOUSES......Page 293
11.6.1 DIFFERENT TYPES OF QUERIES CAN BE ANSWERED AT DIFFERENT LEVELS......Page 294
11.6.2.1 Aggregation semantics for classification rules......Page 295
11.6.2.2 Aggregation semantics for association rules......Page 296
11.6.2.4 Different assumptions or heuristics may be needed at different levels......Page 297
11.7 NONMONOTONIC REASONING IN DATA WAREHOUSING ENVIRONMENT......Page 298
11.8.1 AN ARCHITECTURE COMBINING OLAP AND DATA MINING......Page 299
11.8.2.1 On the use and reuse of intensional historical data......Page 300
11.8.2.2 How data mining can benefit OLAP......Page 301
11.8.2.3 OLAP-enriched data mining......Page 302
11.9.1 MATERIALIZED VIEWS AND INTENSIONAL ANSWERING......Page 304
11.9.2 REWRITING CONCEPTUAL QUERY USING MATERIALIZED VIEWS......Page 305
11.10.1 BASIC APPROACHES FOR WEB MINING......Page 306
11.10.2 DISCOVERY TECHNIQUES ON WEB TRANSACTIONS......Page 307
REFERENCES......Page 309
12.1 OVERVIEW......Page 312
12.2.1 LOGIC AND UNCERTAINTY......Page 313
12.2.2 DIFFERENT TYPES OF UNCERTAINTY AND ONTOLOGIES OF UNCERTAINTY......Page 314
12.2.3 UNCERTAINTY AND SEARCH......Page 315
12.3.1 BASICS OF PROBABILITY THEORY......Page 316
12.3.2 BAYESIAN APPROACH......Page 317
12.3.3.2 Some key concepts in Bayesian networks......Page 318
12.3.3.4 Implementing Bayesian networks......Page 320
12.3.3.5 The notion of d-separation......Page 321
12.3.4.2 An agent-based model for data mining using Bayesian networks......Page 322
12.3.4.3 An example......Page 324
12.3.5 A BRIEF REMARK ON INFLUENCE DIAGRAM AND DECISION THEORY......Page 325
12.3.6.2 Dempster-Shafer Theory......Page 326
12.4.1.1 Probability reasoning versus fuzzy reasoning......Page 329
12.4.1.3 Characteristic functions of fuzzy sets......Page 330
12.4.1.4 Fuzzy decision making systems......Page 331
12.4.2.1 Basic operations......Page 332
12.4.2.2 Triangular norms......Page 333
12.4.3.3 An example......Page 334
12.5.1 FUZZY RELATIONS......Page 335
12.5.2.1 Fuzzy system components......Page 336
12.5.2.3 Fuzzy inference and fuzzy relations......Page 337
12.5.2.4 Fuzzy implication......Page 338
12.5.3.3 Fuzzy rule evaluation......Page 339
12.6 USING FUZZYCLIPS......Page 341
12.7.1 BASICS OF FUZZY CONTROLLER......Page 343
12.7.2 BUILDING FUZZY CONTROLLER USING FUZZYCLIPS......Page 344
12.7.3 FUZZY CONTROLLER DESIGN PROCESS......Page 347
12.8 THE NATURE OF FUZZY LOGIC......Page 350
12.8.2 WHY FUZZY LOGIC HAS BEEN SUCCESSFUL IN EXPERT SYSTEMS......Page 351
SUMMARY......Page 352
REFERENCES......Page 353
13.2.1 REDUCTION AND RECONSTRUCTION ASPECTS IN FUZZY SET THEORY......Page 355
13.2.2 RECONSTRUCTION AND DATA MINING......Page 356
13.3.1 RECONSTRUCTABILITY ANALYSIS USING K-SYSTEMS THEORY......Page 357
13.3.2 REDUCTION-DRIVEN APPROACH IN ROUGH SET THEORY......Page 358
13.4.1 BASIC IDEA OF ROUGH SETS......Page 359
13.4.2 TERMINOLOGY......Page 360
13.4.3 AN EXAMPLE......Page 361
13.4.4 RULE INDUCTION USING ROUGH SET APPROACH......Page 363
13.4.5 APPLICATIONS OF ROUGH SETS......Page 364
13.5 K-SYSTEMS THEORY......Page 365
SUMMARY......Page 367
REFERENCES......Page 368
14.2 INTEGRATED PROBLEM SOLVING......Page 370
14.3.2.1 Solving problems by analogical reasoning......Page 372
14.3.2.4 Solving a problem directly using perturbation......Page 373
14.3.2.7 Inverse problems......Page 374
14.3.2.8 Storage versus recomputation......Page 375
14.3.2.9 Step-wise refinement and manipulation of changes......Page 376
14.4.1.2 Meta-databases......Page 377
14.4.1.3 Meta-searching in the Internet......Page 378
14.4.2.1 General remarks......Page 379
14.4.2.2 Meta-level reasoning for flexible inference control using fuzzy logic......Page 380
14.4.2.3 Combining creativity and expertise using a meta-level interpreter......Page 383
14.4.3.1 Meta-queries and meta rules......Page 384
14.3.3.2 Template design for mining association rules......Page 385
14.4.5 SUMMARY AND REMARK ON META-ISSUES......Page 386
SELF-EXAMINATION QUESTIONS......Page 387
REFERENCES......Page 388