<<

. 136
( 137 .)



>>

uses for, 52

clustering, 57

time
discussed, 7

attributes, market based
uniform distribution, statistics, 132

analysis, 293

uniform product code (UPC), 555

and dates, interval variables, 551

UNIT_MASTER file, customer

dependency, prospecting and, 160





Y
signatures, 559

frames, customer signatures, 562

unordered lists, 239

series analysis





FL
unsupervised learning, 57

neural networks, 244“247

untruthful learning sources, 44“48

non-time series data, 246

AM
UPC (uniform product code), 555

SQL data, 572“573

UPS, transaction processing

statistics, 128“129

systems, 3“4
training sets

up-selling

coverage of values, 232

TE

customer relationships, 467

MBR (memory-based reasoning),

marketing campaigns, 111, 115“116

263“264

U.S. Census Bureau Web site, 94

model sets, partitioning, 71

usage stimulation marketing

optimization as, 230

campaigns, 111

uses for, 52

user roles, data transformation, 58“60

transaction data, OLAP, 476“477
transaction processing systems,
V

customer relationship
validation

management, 3“4
assumptions, 67

transactional records, 574

neural networks, 218

transactors, behavior-based

validation sets

variables, 580

model sets, partitioning, 71

transfer function, neural

test sets, partitioning, 71

networks, 223

uses for, 52

TRANS_MASTER file, customer

value added-services, predication
signatures, 559

tasks, 10

traveling salesman problem, graph

valued outcomes, estimation, 9

theory, 327“329

values

trends, capturing, 75

comparing with descriptions, 65

triangle inequality, distance

with meaning, data correction, 74

function, 272

missing, 590“591

trivial rules, association rules, 297

truncated data, statistics, 162




Team-Fly®
Index 643


Web servers

variables
cookies, 109

data selection, 63“64
transaction processing systems, 3

variable selection problems, neural
Web sites

networks, 233

customer response to marketing

variance

campaigns, tracking, 109

analysis of, 124

RuleQuest, 190

defined, 81

U.S. Census Bureau, 94

neural networks, 199

weight columns, 548

reduction in, splits, decision trees, 183

weighted graphs, graph theory,

standard deviation and, 138

322, 324

statistics, 138

weighted voting, 281“282

variations, percent, 105

weighting, automatic cluster

vectors, angles between, 361“362

detection, 363“365

vendor credibility, 537

welcome periods, loyalty

virtual items, association rules, 307

programs, 518

virtuous cycle

well-defined distance, distance

action tasks, 30

function, 271

business opportunities, identifying,

winback approach, customer

27“28

relationships, 470

data transformation, 28“30

wireless communications industries

discussed, 28

business opportunities, identifying,

results, measuring, 30“32

34“35

stages of, 26

MOU (minutes of use), 38

visualization tools, data exploration, 65

rate plans, finding appropriate, 7

voice node, fax machines, 341

women, differential response analysis

voice recognition, free text

and, 107

resources, 556

word-of-mouth advertising, 283

voluntary churn, 118“119, 521

voting

Z
positive ratings, 284

zip codes

<<

. 136
( 137 .)



>>