Index
Note: Page numbers followed by f indicate figures.
A
Abstraction 144
Accuracy 207–208
Apriori algorithm 221
ASCII editor 185
lexical parsing 27–28
12 lines of Python code 31–34
medical nomenclature 24
natural language autocoders 25–26
nomenclature coding 24–25
on-the-fly autocoding 28
B
Bayesian analysis 297–301
analysis 5
data preparation 4
data structure and content 3
definition 1–2
goals 3
introspection 5
location 3
longevity 4
measurements 4
mechanisms 5–6
purpose of 7–8
reproducibility 4
research universe 8–13
stakes 4
analytic algorithm 332
ASCII editor 185
estimation-only analyses 266
mean-field averaging 267
complete and representative data 188
approximate/local solutions unacceptable 327
incremental 328
model for reality 328
random intervals 327
simple design 327
data description 331
data flattening 200–205
data objects identification and classification 187
data distribution 190
Gnuplot 190
Matplotlib 190
normal/Gaussian distribution 191–192
annotate with metadata 193
data within data object 195
immutable data 196
introspective data 196
membership in defined class 196
scientific value 193
simplified data 197
time stamped data 194
uniqueness/identity 193
data reduction 332
denominators 259–260
formulated questions 329
immutability See Immutability
large files, view and search 198–200
multimodality 270
catchment population 186–187
data manager 186
sample number/dimension dichotomy 187
outliers and anomalies 264–266
preference prediction 268–270
query output adequacy 330
readme/index file 186
data entry errors 426
identification errors 426
medical errors 427
motor vehicle accidents 427
rocket launch errors 427
reformulated questions 330
Big Data designers 428
Big Data indexers 429
data curators 430
data managers 430
domain experts 429
metadata experts 429
network specialists 431
ontologists and classification experts 429
security experts 431
software programmers 429
resource evaluation 329
data analysts 431
data reduction specialists 433
data validators 431
data visualizers 433
free-lance Big Data consultants 434
generalist problem solver 432
scientists with minimal programming skills 433
results and conclusions 335
security policy/restricted data 197–198
self-descriptive information 188
solution estimation 192
validation 336
word frequency distributions 260–264
biomakers 307
cancel-out hypothesis 308
creating unbiased models 303–304
credent results 305–306
death certificates 307–308
DNA sequences 309
hypothesis 304–305
overfitting 309
ambiguity of system elements 313
blending bias 312
complexity bias 313
misguided data 311
statistical method bias 313
Simpson’s paradox 310–311
time-window bias 306–307
Biomakers 307
Black holes 271–274
conditions 177
creating 177–178
properties 178
time stamp 179
triples 177
Burrows Wheeler transform (BWT) 36–50
C
Cancel-out hypothesis 308
Cancer Biomedical Informatics Grid (CaBigTM) 339–344
Class blending 110–111
Class hierarchy 103–104
Classification 101–104
Classifier algorithm 247
drawbacks 246
operation 246
purpose 245
CODIS (Combined DNA Index System) 368–369
Compliance 164–165
Python’s Scipy 243
gene 224–225
medical error/counting errors 212–213
negations 214
systematic counting error 211
word counting rules 211–213
Cryptography 381–387
Cygwin 199–200
D
classification 249–250
drawbacks 246
operation 246
purpose 245
Python’s Scipy 243
data persistence methods 247–249
addition and multiplication 238
cryptographic programs, beware of 241–242
inexact answers 242
one-way hashes 238–240
pseudorandom number generator 240–241
random access to files 237
time stamps 238
NoSQL databases 252–256
random number generator See Random number generator
combinatorics 237
high-speed programming languages 232
iterative loops, system calls within 234
line-by-line reading 233
look-up tables and pre-computed pointers 235
pay for smart speed 237
persistent data 233
proprietary software 234
RegEx language 236
software testing on data subset 233
solutions 231–232
turn-key application 234
unpredictable software 236
utilities 234
SQLite 251–252
advantages 53–55
data objects, naming 55
data scrubbing 69–71
deidentification 66–68
description 53–54
identifier system, properties of 55–58
in image header 71–74
one-way hash 74–82
accession number 63
names 60–61
social security number 62
reidentification 68–69
life science identifiers 64
object identifier 64–66
properties 58
registries 63–64
UUID 58–59
Data Quality Act 397
Data range 209–211
additional analyses and updating results 356
clarification and improved earlier studies 355
data and data documentation errors 353
data misinterpretation 353
data verification 354–355
extending original study 356
irreproducible results 351–352
JADE collider data 356–357
message framing 353–354
outright fraud 353
scientific misconduct 354
validation 355
vindication 357
abandoned data 365–366
CODIS (Combined DNA Index System) 368–369
dark data 365–367
Hadley data 369
legacy data 366–367
novel data sets creation 365
original research performance 364
Plate Boundary Observatory data 369–370
zip codes 367–368
Data scrubbing 69–71
no-cost solution 384
personal identifiers 388–391
algorithm 382
limitations 384
for RSA encryption 382–383
signature and authentication 383–384
use 382
redundancy 385–386
time and money 386
bureaucratic hurdles 380
comply rules 376
data compartmentalization 379
data hackers 378
data misinterpretation 374
data protector 375
flawed data 377
legal ownership 376
limited access to responsible professionals 375
missing data 380
reimbursement 377
research parasites 374
research protocols 379
universal data standards 375
Labeled-Release data on life on mars 387–388
reasons 373–374
Denominators 259–260
Digital Millennium Copyright Act of 1998 (DMCA) 399
Dublin Core 93–95
E
Encapsulation 143
F
abandonware 338
approach to Big Data 328–337
Big Data projects 323
Cancer Biomedical Informatics Grid 339–344
categories 322
data managers 322
Ada 95 323–324
BLOB 323
data management principles 326
instability 325
metric system 324
OSI 323
triples 326
Gaussian copula function 344–347
hospital informatics 322
occurrence 321–322
legacy data, preserving 339
utilities 338
random intervals 327
quantitative data 260
most frequent word 261–262
Pareto’s principle 260
“stop” word 261–262
G
Gaussian copula function 344–347
Gnuplot 190
GraphViz 120–122
H
Hadley data 369
I
ImageMagick 71–72
blockchains and distributed ledgers 176–179
coping with data 173–174
immortal data objects 173
metadata tags 170–171
reconciliation across institutions 174–175
replicative annotations 171–172
trusted timestamp 176
zero-knowledge reconciliation 179–183
Inheritance 143
data object 140–142
abstraction 144
benefits 144
encapsulation 143
feature 139–140
inheritance 143
objects 138
polymorphism 143–144
reflection 144
Ruby 138–139
time stamping 145–147
triplestore 147–152
J
JADE collider data 356–357
L
Labeled-Release study 387–388
accuracy and legitimacy 395–397
biases by consent process 408
confidential consent status 407
confidentiality 404–405
confidentiality risk 409
data managers 404
divert responsibility 410
legally valid consent form 405
preserving consent 407
privacy 405
records of actions 408
retraction 408
train staff on consent-related issues 408
unintended purposes 410
unmerited revenue source 409
Havasupai tribe 413–416
privacy policies 411–412
breaches 402–403
identification theft 403–404
tort 402
copyright laws 398–399
data managers, suggestions for 399–400
Digital Millennium Copyright Act of 1998 399
No Electronic Theft Act of 1997 399
intellectual property 401
license fee 400–401
precautions using 401–402
timely access to data 412–413
unconsented data 409–411
Life science identifiers (LSID) 64
Lotsa data 2
M
Matplotlib 190
accuracy 207–208
biometrics 225–226
control concept 222–223
gene 224–225
medical error/counting errors 212–213
negations 214
systematic counting error 211
word counting rules 211–213
data range 209–211
Apriori algorithm 221
gravitational forces 219
process 221
randomness 220–221
redundancy 219–220
narrow range data 226
population difference, adjusting 216
weighting 218
precision 207–208
statistical significance 223–224
steganography 208
Message digest version 5 (md5) algorithm 74–75
concept 85
Dublin Core 93–95
namespace 88–90
semantics 87–88
XML 85–87
Monte Carlo simulations 288–291
Monty Hall problem 295–297
Mutability See Immutability and identifiers
N
Namespace 88–90
Natural language autocoders 25–26
No Electronic Theft Act of 1997 (NET Act) 399
Noisy class 110–111
Nomenclature coding 24–25
NoSQL databases 252–256
O
Object by relationships 97–101
Object by similarity 98–100
creating 64–65
HL7 65–66
problem 65
Object oriented programming 116
abstraction 144
benefits 144
data object, assigning 107
encapsulation 143
feature 139–140
inheritance 143
multiclass inheritance 107
objects 138
polymorphism 143–144
reflection 144
Ruby 138–139
syntax 106–107
On-the-fly autocoding 28
class blending (noisy class) 110–111
Aristotle 102
biological classifications 101–102
data domain 104
data objects hierarchy 102
parent class 103–104
taxonomy 104
Big Data resource 108–109
complex ontology 108–109
inheritance rules 107
multiclass inheritance 107–108
object oriented programming 106–107
Python/Perl programming languages 106
Ruby programming language 106
simple classification 109
GraphViz 120
RDF Schema 123–124
multiple parent classes 104–106
paradoxes 115–116
classes and properties 113
descriptive language 113
miscellaneous classes 112
transitive classes 112
RDF Schema 117–120
upper level ontology 114–115
Outliers 264–266
Overfitting 309
P
Personal identifiers 388–391
Plate Boundary Observatory data 369–370
Polymorphism 143–144
Precision 207–208
Pseudorandom number generator 240–241
sample 278
simple simulation 278–279
Python/Perl programming languages 106–107
Python’s Scipy 243
R
Bayesian analysis 297–301
frequency of unlikely occurrences 293–294
Monte Carlo simulations 288–291
Monty Hall problem 295–297
sample 278
simple simulation 278–279
output/conclusion 287
power estimates 288
random numbers generation 285–286
repeated simulation 286–287
sample size 288
scalability 287–288
shuffling 284–285
statistical method 284
Reflection 144
and class properties 117–120
features 90
GraphViz 123–124
syntax for triples 91–92
S
Semantics 87–88
Simpson’s paradox 310–311
Small data 3–5
anti-hypothesis 421
Big Brother hypothesis 420
Big Snoop hypothesis 419
Borg invasion hypothesis 420
decision-making algorithms 425–428
Egghead heaven hypothesis 421
Facebook hypothesis 421
George Orwell’s 1984 440–442
hubris and hyperbole 434–437
Junkyard hypothesis 420
public mistrust 424–425
reduced cost and increased productivity 422–424
resource builders 428
resource users 431
Scavenger hunt hypothesis 421
complex 161
compliance 164
strength 161
SQLite 251–252
Chocolate Teapot 165–167
coercion 162
complex 161
compliance 164–165
construction rules 160–161
creation 157
Darwinian struggle 162
filtering-out process 156
measures 162
new standards 156
popular 159
profit 157
purpose of 159
strength 161
versioning 161–164
Suggested Upper Merged Ontology (SUMO) 114–115
T
Term extraction 19–22
Time-window bias 306–307
Triplestore 147–152
Trusted timestamp 176
U
life science identifiers 64
object identifier 64–66
properties 58
registries 63–64
UUID 58–59
collisions 59
Linux 59
properties 58
Python 59
V
Versioning 161–164
W
Word counting 211–213
World Intellectual Property Organization (WIPO) 401
X
drawback 86–87
importance 86
properties 86
syntax 85–86
XML Schema 86
Z
Zero-knowledge reconciliation 179–183
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.