intuition, data exploration, 65 involuntary churn, 118-119, 521 item popularity, market based

analysis, 293 item sets, market based analysis, 289 Iterative Dichotomiser 3 (ID3), 190

key and ID variables, 554 KDD (knowledge discovery in

databases), 8 Kimball, Ralph (The Data Warehouse

Toolkit), 474 Kleinberg algorithm, link analysis,

332-333 K-means clustering, 354-358 knowledge discovery in databases

(KDD), 8

Kolmogorov-Smirnov (KS) tests, 101 L

large-business relationships, customer

relationship management, 3-4 leaf nodes, classification, 167 learning

opportunities, customer interactions,

520-521 supervised, 57 training techniques as, 231 truthful sources, 48-50 unsupervised, 57 untruthful sources, 44-48 life stages, customer relationships, 455-456

lifetime customer value, customer

relationships, 32 lift ratio

comparing models using, 81-82

lift charts, 82, 84

problems with, 83 linear processes, 55 linear regression, 139 link analysis

authorities, 333-334

candidates, 333

case study, 343-346

classification, 9

discussed, 321

fax machines, 337-341

graphs

acyclic graphs, 331

communities of interest, 346

cyclic, 330-331

data as, 340

directed graphs, 330

edges, 322

graph-coloring algorithm, 340-341 Hamiltonian path, 328 nodes, 322 planar graphs, 323 traveling salesman problem,

327-329 vertices, 322 hubs, 332-334

Kleinberg algorithm, 332-333

root sets, 333

search programs, 331

stemming, 333

weighted graphs, 322, 324 linkage graphs, 77 lists, ordered and unordered, 239 literature, market research, 22 logarithms, data transformation, 74 logical schema, OLAP, 478 logistic methods, box diagrams, 200 long form, census data, 94 long-term trends, 75 lookup tables, auxiliary information,

570-571 loyalty

customers, 520

loyalty programs marketing campaigns, 111 welcome periods, 518 luminosity, 351

mailings marketing campaigns, 97 non-response models, 35

marginal customers, 553 market based analysis

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 market research

control group response versus, 38

literature, 22

shortcomings, 25

survey-based, 113 marketing campaigns. See also advertising

acquisitions-time data, 108-110

canonical measurements, 31

champion-challenger approach, 139

credit risks, reducing exposure to, 113-114

cross-selling, 115-116

customer response, tracking, 109

customer segmentation, 111-113

differential response analysis, 107-108

discussed, 95

fixed budgets, 97-100

loyalty programs, 111

new customer information, gathering, 109-110

people most influenced by, 106-107

planning, 27

profitability, 100-104

proof-of-concept projects, 600

response modeling, 96-97

as statistical analysis acuity of testing, 147-148 confidence intervals, 146 proportion, standard error of, 139-141

results, comparing, using confidence bounds, 141-143 sample sizes, 145 targeted acquisition campaigns, 31 types of, 111 up-selling, 115-116 usage stimulation, 111 marriages categorical values, 239-240 house-hold level data, 96 mass intimacy, customer relationships,

451-153 massively parallel processor

(MPP), 485 maximum values, of simple functions,

generic algorithms, 424 MBR. See memory-based reasoning MDL (minimum description

length), 78 mean between time failure

(MTBF), 384 mean time to failure (MTTF), 384 mean values, statistics, 137 measurement errors, 159 median customer lifetime value,

retention, 387 median values, statistics, 137 medical insurance claims, useful

data sources, 60 medical treatment applications,

MBR, 258 meetings, brainstorming, 37 memory-based reasoning (MBR) case study, 259-262 challenges of, 262-265 classification codes, 266, 273-274 combination function, 258, 265 customer classification, 90-91 customer response prediction, 258

memory-based reasoning (MBR) (continued)

democracy approach, 279-281

distance function, 258, 265, 271-272

fraud detection, 258

free text response, 258

historical records, selecting, 262-263

medical treatment applications, 258

new customers, 277

relevance feedback, 267-268

similarity measurements, 271-272

training data, 263-264

weighted voting, 281-282 men, differential response analysis

and, 107 messages, prospecting, 89-90 metadata repository, 484, 491 methodologies

data correction, 72-74

data exploration, 64-68

data mining process, 54-55

data selection, 60-64

data transformation, 74-76

data translation, 56-60

learning sources truthful, 48-50 untruthful, 44-48

model assessment, 78-82

model building, 77

model deployment, 84-85

model sets, creating, 68-72

reasons for, 44

results, assessing, 85 metropolitan statistical area (MSA), 94 minimum description length

(MDL), 78 minimum support pruning, decision

trees, 312 minutes of use (MOU), wireless

communications industries, 38 misclassification rates, binary classification, 98

missing data data correction, 73-74 NULL values, 590 splits, decision trees, 174-175 mission-critical applications, 32 mode values, statistics, 137 models assessing classifiers and predictors, 79 descriptive models, 78 directed models, 78-79 estimators, 79-81

building, 8, 77

comparing, using lift ratio, 81-82 deploying, 84-85 model sets balanced datasets, 68 components of, 52 customer signatures, assembling, 68 partitioning, 71-72 predictive models, 70-71 timelines, multiple, 70 non-response, mass mailings, 35 score sets, 52 motor vehicle registration records,

useful data sources, 61 MOU (minutes of use), wireless

communications industries, 38 MPP (massively parallel processor), 485 MSA (metropolitan statistical area), 94 MTBF (mean between time failure), 384 MTTF (mean time to failure), 384 multiway splits, decision trees, 171 mutation, generic algorithms, 431-132

N variables, dimension, 352 National Consumer Assets Group

(NCAG), 23 natural association, automatic cluster

detection, 358

Team-Fly®