Calibrating crime, 22–24
Calls for service (CFS), 47
Card-not-present fraud, 277–78
Case Notebook, 90
Case studies
bio-surveillance, 117–20
border machine learning, 210–12
Chicago Internal Affairs, 167–68
ChoicePoint, 66–68
chromatogram retrieval system, 172–77
clustering burglars, 24–37
death claims, 287–88
drug trafficking, 81–82
government financial transaction detection, 213–19
intrusion detection, 313–18
military data extrapolation, 212
MO modeling, 179–95
money laundering, 84–85
sexual assault offender behavior modeling, 348–62
wireless fraud, 78–79
CATCH, 364–75
abstract, 365
analysis tools, 366
ANNs, 365, 366
clustering algorithm, 366–67
clustering tools, 365
conclusion, 374
crime data vectors, 370
crime similarity, viewing, 371
crime type comparison, 372
database mining, 367–70
database visualization, 370–73
defined, 365
evaluation, 373–74
introduction, 365–66
order and description of crimes, 369
potential, 374
SOMs, 366, 367
SQL query generation, 369
starmap of crimes, 368
tool types, 367
versions, 366
Center of Excellence in Document Analysis and Recognition (CEDAR) project, 363–64
defined, 363
information on, 364
Charts
clarity, 92–93
confirmed/unconfirmed lines, 91
high volume data, 95
legend, 94
link types, 93
organizations inside boxes, 91–92
telephone toll analysis, 94
See also Analyst's Notebook
Chicago Internal Affairs case study, 167–68
Chi-square automatic interaction detection (CHAID), 12, 205
design, 206
operation, 12
See also Machine-learning algorithms
ChoicePoint case study, 66–68
AutoTrack Wireless, 66
CORE service, 68
defined, 66
National Comprehensive Report, 67–68
notification, 68
SQL Direct, 66–67
Chromatogram retrieval system case study, 172–77
Clairvoyance, 142–43
applications, 143
defined, 142
functions, 142–43
See also Text mining tools
Classification, 159
Classification and regression trees (CART), 205, 224–27
accuracy, 225
binary trees, 226
DBMS/COPY, 225
defined, 224
design, 206
IF/THEN rule generation, 224
interface, 225
rules generation, 227
rules sample, 228
See also Decision tree tools; machine-learning algorithms
Classifier version 5 (C5.0), 12, 205
classifiers, 224
defined, 206–7, 224
See also Machine-learning algorithms
ClearForest, 143–44
defined, 143
dynamic view of relationships, 144
output, 144
taxonomy graphical view, 143
See also Text mining tools
Clementine, 26, 30, 179, 245–47
application templates (CATs), 245–47
COP CAT, 245–46
defined, 245
government fraud CAT, 246
police computer systems interface, 181
SOM option, 31
See also Machine-learning software; SPSS
Clustering, 159–60
border smugglers with SOM, 169–72
fraud, 262–63
Clustering burglars case study, 24–37
application construction, 30–31
application task, 26–27
conclusions, 36
data selection, cleaning, coding, 27–30
discussion, 34–36
findings, 31
introduction, 25–26
room for improvement, 34–35
validation process, 31–33
Clusters
analysis, 31–32
crime, 346–47
dendrograms for, 35
COGNOS 4Though, 197
Cognos Scenario, 227–28
Combined DNA Index System (CODIS), 52
Computer-aided dispatch (CAD) systems, 343
Computer Aided Tracking and Characterization of Homicides and Sexual Assaults. See CATCH
Conditional rules, 21
Consolidation, 78
Cookies, 57–58
anatomy, 58
defined, 57
example, 57–58
See also Internet data
Copernic, 144–45
Credit-card fraud, 250–51
detection technique, 277
MO, 277
Credit data, 46–47
Crime clusters, 346–47
Crime detection
data balancing, 296
data mining methodology, 293–96
data rotation, 297
data splitting, 296–97
deployment and monitoring, 298–99
false positive measurements, 298
mechanisms, 296–99
model combining, 297–98
multiple model evaluation, 297
random sampling, 296
Crime Link, 97–98
defined, 97
diagram generation, 98, 99
two-dimensional association matrix, 97–98
See also Link analysis tools
Crime maps, 343–45
defined, 343–44
illustrated, 345
interactive, 345–46
persuasive nature of, 344–45
Crime(s)
bogus official, 25, 26, 27
calibrating, 22–24
by day of week, 188
detecting, through data mining, 220–21
detection methodology, 293–96
digital, 219–20
financial, 277–78
identity, 291–92
insurance, 281–87
mapping, 343–78, 370
probability of, 23
property, 18
rate, computing, 16–18
reports, 27
similarity of, 371
targets of, 18
telecommunications, 288–90
by time of day, 188
type comparison, 372
Crime Workbench, 98–100
Action Management module, 98
database searches, 99
defined, 98
Link Management module, 99
Web, 99–100
See also Link analysis tools
Criminal analysis
comparison crime rates, 16–18
crime pattern breakdown, 16
data mining and, 15–19
environment and, 16
goal, 16
Criminal data, 47–55
aggregate statistics, 49
CFFS, 47
CFS, 47
CJIS WAN, 52
CODIS, 52
DRUGX, 53
FinCEN, 53
GREAT, 54
IDENT-INS, 53
JABS, 53
LEO, 53
NCIC, 51–52
NCVS, 48
NDPIX, 53
NIBIN, 52
NIBRS, 47, 48–49
NICS, 53
NIPC, 53
NLETS, 50–51
RISS, 53
SWBS ADIS, 53
TECS, 53
UCR/NIBRS, 53
Web sites, 54–55
WIN, 53
See also Data
Criminal Justice Information Services Wide Area Network (CJIS WAN), 52
Criminal profiling. See Profiling
CRISP-DM, 180, 293–96, 349
concept, 293–94
data preparation, 294–95
data understanding, 294
defined, 293
deployment, 295–96
diagram arrows, 293
evaluation, 295
investigation objective understanding, 294
modeling, 295
objective, 294
phases, 293
processes, 294–96
Cron, 308
CRoss-Industry Standard Process-Data Mining. See CRISP-DM
Customer-relationship management (CRM), 41
Cybercrimes, 301–2
CyberSource, 271