Index

absolute values, distance function, 275 accuracy

classification and predication, 79

estimation, 79-81 acquisition

acquisition-time data, 108-110

customer relationships, 461-164 actions

actionable data, 516

actionable results, 22

actionable rules, association rules, 296

control group response versus market research response, 38

taking control of, 30 activation function, neural

networks, 222 acuity of testing, statistical analysis,

147-148 ad hoc questions

behavior-based variables, 585

business opportunities, identifying, 27

hypothesis testing, 50-51 additive facts, OLAP, 501 addresses, geographical resources, 555-556

adjusted error rates, CART

algorithm, 185 advertising. See also marketing campaigns

communication channels, prospecting, 89

prospects, 90-94

word-of-mouth, 283 affinity grouping

association rules, 11

business goals, formulating, 605

cross-selling opportunities, 11

data transformation, 57

undirected data mining, 57 affordability, server platforms, 13 agglomerative clustering, automatic

cluster detection, 368-370 aggregation, confusion and, 48 aggression, behavior-based

variables, 18 AI (artificial intelligence), 15 algorithms, recursive, 173 alphas, decision trees, 188 American Express

as information broker, 16

orders, market based analysis, 292

analysis differential response, 107-108 link analysis

acyclic graphs, 331

authorities, 333-334

candidates, 333

case study, 343-346

classification, 9

communities of interest, graphs, 346 cyclic graphs, 330-331 data, as graphs, 340 directed graphs, 330 discussed, 321 edges, graphs, 322 fax machines, 337-341 graph-coloring algorithm, 340-341 Hamiltonian path, graphs, 328 hubs, 332-334

Kleinberg algorithm, 332-333

nodes, graphs, 322

planar graphs, 323

root sets, 333

search programs, 331

stemming, 333

traveling salesman problem,

graphs, 327-329 vertices, graphs, 322 weighted graphs, 322, 324 market based differentiation, 289 discussed, 287 geographic attributes, 293 item popularity, 293 item sets, 289

market basket data, 51, 289-291 marketing interventions, tracking,

293-294 order characteristics, 292 products, clustering by usage,

294-295 purchases, 289 support, 301

telecommunications customers, 288 time attributes, 293

sensitivity, 247-248 sequential, 318-319 statistical acuity of testing, marketing

campaign approaches, 147-148 business data versus scientific

data, 159 censored data, 161 Central Limit Theorem, 129-130 chi-square tests, 149-153 confidence intervals, marketing

campaign approaches, 146 continuous variables, 137-138 correlation ranges, 139 cross-tabulations, 136 density function, 133 as disciplinary technique, 123 discrete values, 127-131 experimentation, 160-161 field values, 128 histograms and, 127 mean values, 137 median values, 137 mode values, 137 multiple comparisons, 148-149 normal distribution, 130-132 null hypothesis and, 125-126 probabilities, 133-135 proportion, standard error of,

marketing campaign

approaches, 139-141 p-values, 126 q-values, 126 range values, 137 regression ranges, 139 sample sizes, marketing campaign

approaches, 145 sample variation, 129 standard deviation, 132, 138 standardized values, 129-133 sum of values, 137-138 time series analysis, 128-129 truncated data, 162

variance, 138 z-values, 131, 138 survival

attrition, handling different types

of, 412-113 customer relationships, 413-115 estimation tasks, 10 forecasting, 415-116 time series neural networks, 244-247 non-time series data, 246 SQL data, 572-573 statistics, 128-129 of variance, 124 analysts, responsibilities of, 492-493 analytic efforts, wasted time, 27 AND value, neural networks, 222 angles, between vectors, 361-362 anonymous versus identified

transactions, association rules, 308 application programming interface

(API), 535 architecture, data mining, 528-532 artificial intelligence (AI), 15 assessing models classifiers and predictors, 79 descriptive models, 78 directed models, 78-79 estimators, 79-81 association rules actionable rules, 296 affinity grouping, 11 anonymous versus identified

transactions, 308 data quality, 308 dissociation rules, 317 effectiveness of, 299-301 inexplicable rules, 297-298 point-of-sale data, 288 practical limits, overcoming, 311-313 prediction, 70

probabilities, calculating, 309 products, hierarchical categories, 305

sequential analysis, 318-319

for store comparisons, 315-316

trivial rules, 297

virtual items, 307 assumptions, validation, 67 attrition

discussed, 17

forced, 118

future, 49

proof-of-concept projects, 599

survival analysis, 412-413 audio, binary data, 557 authorities, link analysis, 333-334 automated systems

neural networks, 213

transaction processing systems, 3-4 automatic cluster detection

agglomerative clustering, 368-370

case study, 374-378

categorical variables, 359

centroid distance, 369

complete linkage, 369

data preparation, 363-365

dimension, 352

directed clustering, 372

discussed, 12, 91, 351

distance and similarity, 359-363

divisive clustering, 371-372

evaluation, 372-373

Gaussian mixture model, 366-367

geometric distance, 360-361

hard clustering, 367

Hertzsprung-Russell diagram, 352-354

K-means algorithm, 354-358

luminosity, 351

natural association, 358

scaling, 363-364

single linkage, 369

soft clustering, 367

SOM (self-organizing map), 372

vectors, angles between, 361-362

weighting, 363-365

zone boundaries, adjusting, 380