Index
NOTE: An italicized f, ff, and t following a page number denotes a figure, multiple figures, and tables, respectively, on that page.
A
Abduction, child, 210–211
Actionable Mining model, 99–100
Ad hoc databases, 74–76
Alert fatigue, 229
Algorithms
applied to data, 60–61
modeling, 51
predictive, 29
selecting, 125–126
unsupervised learning, 169
Al Qaeda, 269
Analysis. See also Link analysis sequential iterations of, 46–47
Analysis of Violent Crime, National Center
for the, 210
Analysts
fieldwork, 20–22
operational personnel and, 22–24
rapid analysis, 75
Analytical stream, sample, 136f
Analytics, predictive, 28
Animal research, 189–191
Antisthenes, 184
Apollo 13 (movie), 113–114
Arrest-based crime reporting, 147
Arrest data, 13–14
Assaults
among drug sellers, 202
robbery-related, 7
sexual, 206–208
Automated classification techniques, 203
Automated motive determination, 203–205
B
Baseline data, 12–14
Bayes’ Theorem, 233
Behavior
criminal, 177
deviations from, 130
economic motive for crime, 207
effect of weather events on, 91
future, predictions about, 277
identification and analysis of a sample of, 6
“normal” criminal, 181–182
similar, repeated identification of, 98–99
violence, 188
Belsan siege, 224–225
Billing invoice, telephone, 105f–109f
Binary data, 70
Bohr, Neils, 215
Boosting, 311
Brody, Herb, 259
“Brute force” analytics, 28
Business understanding, 50
C
Calls for service data
during analysis, 246–247
challenges associated with analyzing, 81–82
citizen complaints, random gunfire, 80t–81t
overview, 78–79
Cameras, digital, 227
Car theft, 161–162
Cash flow, 274
Categorical data, 93
Census data, 245–246
Central Intelligence Agency
Intelligence Process models
analysis and production phase, 48–49
collection phase, 47–48
dissemination phase, 49
feedback, 49
processing and exploitation phase, 48
requirements phase, 47
Changes, documenting, 99
Chat rooms, 318
Child abduction, 210–211
CIA. See Central Intelligence Agency
CIA Intelligence Process model
overview, 52
support for, 52
Circular logic, 202–203
Classification techniques, automated, 203
Clementine stream, sample, 136f
Clustering, Two-Step, 122
Clustering algorithms, 129f
Coalition Provisional Authority headquarters, 216f
Cognitive Neuroscience, 122–125
Cold case investigation, 194
Collection modalities, 89
Columbo (t.v. show), 194
Communication, 227–228
Community violence, 187
Composite crime indices, 148
Compromise, 22–24
Confidence matrix, 159t
Consistency, 76
Control-charting function, 182
Cops & Docs program, 130
Cost analysis, 229
Craven, John, 233
Crime
benefits of analyzing, violent, 205
economic motive for, 207
evaluation based on homicide rate, 149
normal, 177–178
staged, 183–184
tools used in analysis, 192–193
victimology, 208–209
Crime control strategies, 146
Crime data, 89
Crime displacement, 150–151
Crime prevention, economical viewpoint, 153
Crime reporting, arrest-based, 147
Crimes. See also Assaults; Homicides; Violent crimes
associating to known suspects, 198–199
D.C. sniper investigation, 32–33
drug-related, 203
evenly distributed, 82
frequency, related to cash flow, 274
heat maps, 249–250
identifying frequent, 166
Laci Peterson disappearance, 32
nonadjudicated activity, 86–87
property-related, 181
reduction, targeted approaches to, 31
specialized databases, 74–75
Criminal behavior. See Behavior
Criminals
D.C. sniper, 32–33
drug dealers, 35–36
population samples, 4–6
CRISP-DM process
domain expertise, 52
evaluation, 51
overview, 49–50
phases of, 50–51
D
Data. See also Text data
accurately characterizing, 35–36
calls for service (See Calls for service) continuous, 70
partitioning, 311–312
resources, 71–72
for risk and threat assessment, 227 spatial attributers (See Trinity Sight
analytical model)
Databases
for case management, 72–73
specialized, 74–75
tip, 39–40
Data inventory, 57–58
Data mining
definition, 25
future trends, 40
intuitive nature of, 23–25
translating output, 167
web-based deployment, 168f
Data Mining Moratorium Act (2003), 25
Data mining tools
“black box,” 228
considerations based on needs, 33–34
data mining algorithms, 304
domain-specific tools, 319
for information-based operations, 45
mapping software, 276
specialty niche markets, 34
text mining, 39–40
Data partitioning, 311–312
Decision making, group-based, 232–233
Department of Defense, U.S., 232
Deployment
census data, 245–246
data, 241–242
decisions, 239–240
effects of weather on, 242–245
exploratory graphics, 246–247
goals, 239
homeland security and, 264–265
of information over the Internet, 288
patrol services, 240
of police resources, 239
risk-based case studies, 259–265
schedule, 248f
structuring, 240–241
tactical, 250–252
threat assessment of information, 294
Deployment analysis, 94
Descriptive statistics definition, 4
Diagnostic tool, 132
Digital cameras, 227
Discrete data, 70
Discriminant analysis learning algorithms, 122
Disraeli, Benjamin, 3
Distribution of complaints, 150f
Distributions, uneven, 9–11
DNA
cold hits, for sex crimes, 206
database, 181
offenders identified by, 181
Southside Strangler case, 30–31
Documents, escalation, 99
Domain expertise
for analysts, 20–22
definition, 19–20
deployment strategies for, 246–247
importance of, 52
variable selection process, 59–60
Domain-specific tools, 319
Dorn, Chris, 224
Dorn, Michael, 224
Drive-by shootings. See Victims
Drug arrests, information map, 173f
Drug markets, illegal, 191
changing patterns and trends, 46
rule sets, 29–30
Drug-related violence. See Violence
Duplication, 100
E
Embezzlement, 178–181
Enterprise Miner analytical process, 136f
Enterprise miner analytical process, 136f
Errors in data entry, 83–85
Events
infrequent, 7–8
low-frequency, 217–218
Evidence. See also DNA evidence validity checks on, 87
Expert systems
experts versus, 218–219
lack of bias in, 194–195
F
Facility map with overlay, 170f
Fahrenheit temperature scale, 70
Fear, 227
Federal Bureau of Investigation
National Center for the Analysis of Violent Crime, 210
Uniform Crime Reports, 13
Violent Crime Apprehension Program, 89
Ferrara, Paul, 206
Fieldwork, 20–21
Fight or flight response, 190–191
Firearms
assaults among drug sellers, 202
involvement with, 210
Project Exile, 144–145
sawed-off shotguns, 59
violent offenders and, 210
Flag data, 70
Formatting, 82–83
Functional interoperability, 318
Fusion of multiple resources, 56–57
Futures Markets Applied to Prediction (FutureMAP) program, 232–233
G
Generalizability vs accuracy, 14–17
Giduck, John, 224–225
Goldstein, Paul, 205
Goldstein model, 196
Google, 318
Graphical representations, heat maps, 166
Gunfire
citizen complaints of, 145–146
New Year’s Eve initiative, 151–153
H
Hawaii Five-O (t.v. show), 194
Hierarchical organizational strategies, 97
Hoffer, Eric, 177
Homicide rate, 147
Homicides
categorizing, 46–47
data analysis, 27–28
decision tree models, 196–197
drug-related rule set for, 211
identifying motives, 29
Southside Strangler case, 30–31
victim-perpetrator relationships, 196–197
Hostage siege and massacre, Nord-Ost Theater, 224
Human-Source Intelligence (HUMINT), 48
Hurricane Katrina, 144
Huxley, Thomas Henry, 178
I
Identity theft, 302–303
Illegal drug markets. See Drug markets, illegal
Imagery Intelligence (IMINT), 48
Improvement, indicators of, 85
Imputation, data, 100–101
Incident reports, 55–56
Incidents, nature of, 99–100
Indicator variables, 84
Inferential statistics, 4
Information samples, 5
Information web, 317f
Injury
aggregate analysis of, 210
behavioral styles associated with risk, 35–36
Insider information, 233
Integrated surveillance analytical model, 305
Intelligence Process model, CIA, 47–49. See also Models
Internal rule sets, 29
International Association of Chiefs of Police, 40
Internet activity patterns, 295–296
Internet data, 110–111
Internet honeypots, 293
Internet surveillance, 287
Interoperability, functional, 318
Interval scale, 70
Intrusion detection, 301–302
Investigative efficacy, 211
Iterative processes, 45–47
J
Juvenile delinquency, 95
K
Kohonan network models, 125
L
Law enforcement officer safety, 36 predictive analytics, 28
Learning techniques, 119–121
Likelihood
false positives in infrequent events, 220 of risk, 215
call topography, 120f
identifying relationships in data, 119
interpreting, 37
overview, 9–11
software, 38–39
tool, example, 120f
Link chart, 9–10
Link charts, 10ff
Locations vulnerable to attack, 222
Low-frequency events, 217–218
M
MARGIN. See Mid-Atlantic Regional Gang Investigators Network
McLaughlin, Phillip, 19
Measure, specific, 147–149
Measurement and Signature Intelligence (MASINT), 48
Mid-Atlantic Regional Gang Investigators Network, 89
Millennium bomber, 269
Minority Report (movie), 28
Misrepresentation
emotional victims, 86
nonadjudicated crimes, 86–87
outliers, 87
Missing, 84
Missing data, 84
Models. See also Central Intelligence Agency, Intelligence Process models; specific models
accuracy, 62
in the applied setting, 60
comparisons, 64f
complexity of, 15
deployment, 16
drug-related homicides, 17–18
evaluation phase, 61–62
generalizability vs accuracy, 14–17
lack of a control group in, 221–222
overfitting, 14
overview, 6
purpose of, 153–154
revised, 23
revising and adjusting, 154
rule induction, 61
updating, 161–162
Motive determination, automated, 203–205
Motive determination model, 113
Motives
identifying, 29
predictor of, 113
Multiple resources, 56–57
Murder. See Homicides
N
National Center for the Analysis of Violent Crime, 210
National Incident-Based Reporting System, 77
National Security in the Information Age, 319
Nature, 146
Nature of the incident, 99–100
Network models, Kohonan, 125
Neural nets, 122–123
Neural networks, 123–125
New relationships within, 30–31
NIBRS. See National Incident-Based Reporting System
Nominal scales, 71
Nonobvious relationship analysis (NORA), 37–39
NORA. See Nonobvious relationship analysis (NORA)
Nord-Ost Theater hostage siege and massacre, 224
Normal (term definition), 128–129
Norms, internal, 127,128
O
Offending patterns, 256
Officer safety, 36
Open-Source Information (OSINT), 48
Operational limitations, 21
Operationally actionable output, 63
Operational personnel
analysts and, 22–24
definition, 20
Operational value, 59–60
Ordinal scales, 70
Organizing tips, 32–33
Orthophotography, 172f
Outcome evaluation, 143
Outcome measures, 144–146
“Outlaw lifestyle,” 205
Output
hostile surveillance activity, 231–232
operationally actionable, 63
sample, 168f
Overkill, 193
Overview, 51–52
P
Part I crimes, 148
Pattern recognition, 193–194
Patterns
identifying changes in, 46
identifying patterns of, 207
PDAs, 170f
“Perfect world scenario,” 133–134
Persistent cookies, 111
Police deployment. See Deployment
Police reports, 39
Population, samples versus, 4
Population statistics, 12–13
Predators, sexual, 208
Predictions
armed robbery escalation, 7
bias in, 13
Predictive algorithms, 29
Preparation, 50–51
Prior probabilities, 9
Priors. See Prior probabilities
Privacy issues, 25–27
Probabilities, prior, 138–139
Promiscuity, 210
Property-related crimes, 181
Public health data, 90
R
Racial distribution, 12–13
Random assignment software procedures, 155
Random selection, 154–155
Rapists, stranger, 206–208
Ratio scale, 70
Reality testing, 29–30
RECAP. See Regional Crime Analysis Program
Recoding
about, 39
continuous information, 93
iterative process, 95
offense information, 100
relevant preprocessing, 57–59
spatial, 97
specific dates, 95
Records
duplicate, 88–89
management systems, 72–74
Regional fusion centers, 89
Relationships, victim-perpetrator, 201
Reliability, 83
Reliability check, 85
Request for analysis, 68–69
Rescorla, Rick, 230
Response planning, 230–231
Response rates, 85
Return on investment, 151–153
Reverse lookup programs, 101–102
Risk-based deployment overview, 251–252
Risk-based deployment strategies, 45
Risk evaluation and mitigation, 215
Risk factors, victim, 209–210
ROI. See Return on investment
Rule induction models, 61
S
Samples
behavior, 6
population vs, 4
random composition, 5–6
Schools
cameras in, 227
Chechen attack on, 224
emergency response plans, 229–230
hostile surveillance of, 227–228
preparedness for violence, 230–231
vulnerability of, 222–224
Screening test, 132
Screening tool, 132
Seasonal changes, 161
Self-generated databases, 74–76
Sets, 70
Sexual assault, 206–208
Sexual predators, 208
Shermer, Michael, 232
Signals Intelligence (SIGINT), 48
Signal-to-noise issues, 11
Situational awareness, 165
Snow, Charles Percy, 189
Software products. See Data mining tools
Southside Strangler case, 30–31
Space, 145–146
Spatial boundaries, delineation of, 306
Spatial recoding, 97
Spatial refinement, 231
Specialty niche markets, 34
Species-specific defense reactions, 190
Specific measure, 147–149
Spectators, 85
Spencer, Timothy, 31
Spikes, unexpected, 182
SPSS and Information Builders, 169
SSDR. See Species-specific defense reactions
Statement analysis, 39–40
Statements, value of, 87–88
Statistics, 4
Strategic characterization, 220–222
Strategies, 45
Street maps, 79f
Suicide bombers, profiles of, 220
Surveillance
correlation in detection, 303–304
identifying specific locations, 275–276
important aspect of detection, 232
internet data, 110–111
natural, 270–275
operational benefits, 225–226
operational plans, 286–289
risk and threat assessment, 219–220
schools, 227–228
syndromic, 303
techniques in crime analysis, 268–269
Suspects
circular logic, 202–203
Syndromic surveillance, 303
T
Television station KXTV, 32
Temporal measures, 94
Terrorism
alert fatigue, 229
Chechen attack on a school, 224
cost analysis, 229
deployment of information, 288
geographically distinct attacks, 269–270
Nord-Ost Theater hostage siege and massacre, 224
response planning, 230–231
school emergency response plans, 229–230
September 11th attack, 217
strategic characterization of terrorists, 220–222
tactics, importance of understanding, 224–225
Terrorism Information Awareness system, 26
Test samples, 154–158
Text data, 71
Time blocks, 94
Tip databases, 76
Tools. See Data mining tools
Total Information Awareness system (TIA), 26
Trinity Sight analytical model
nature of the incident, 99–100
space, 97–99
time, 94–97
Two-Step clustering, 122
U
UCR. See Uniform Crime Reports
Uneven distributions, 9–11
Unformatted data, 71
Uniform Crime Reports, 77
U.S. Department of Defense, 232
V
Validity, 83
Variables
continuous, 70
domain expertise to identify, 114
selection, 111–114
spatial, 98
Variable selection, 59
actionable model, 112–113
data quality, 112
VICAP. See Violent Crime Apprehension Program
Victimology, 208–209
Victim-perpetrator relationship, 201
Victim risk factors, 209–210
Victims
attributes related to risk, 209
drive-by shootings, 206
emotional, 86
identification of risk factors, 226
impaired judgment, 95
Violence
global injury prevention programs, 47–48
reduction strategy, 263
risk-based deployment strategies, 260
in schools, preparedness for, 230–231
Violent Crime, National Center for the Analysis of, 210
Violent Crime Apprehension Program, 89
Violent crime index, 148
Violent crimes
behavior analysis of, 198–199
modeled using advanced statistics, 202
Virginia, Commonwealth of, 30–31
Virtual warehouses, 318
Vulnerability, children, 222
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.