Index

@data(), 133

@everywhere, 327

@parallel, 327

@time, 72, 312

abs(), 196, 198

accuracy rate, 39, 167, 209, 210

addprocs(n), 326

alphanumeric operator, 76

ANN. See artificial neural network

ANN2vector(), 263

append!(), 65

apply(), 74, 272

apply_forest(), 257, 259

apply_forest_proba(), 257

apply_tree(), 252

apply_tree_proba(), 252

array, 51, 52–55

array(), 196

artificial neural network, 260–65

association rules extraction, 106

Atom software, 15

AUC metric, 216

AUC(), 216

bar plot, 162

basic accuracy, 208

basic random sampling, 38

Bayesian network, 275

BigFloat, 51, 52

BigInt, 51, 52

binarization, 143

binning, 143

Bokeh, 161, 184

Bool, 51, 61, 96, 144, 145, 147, 290

boosted tree, 273

Botlzman machine, 274

box plot, 162

break, 79

build_forest(), 256, 258

by(), 133

Canopy, 14

centers, 233

Char, 51

chi-square test, 160, 175

classification, 34, 148

classify(), 35, 36

cleaning up data, 137–39

cliques, 293

clustering, 119

CM2EL(), 291

C-means, 231

code

organizing, 25

referencing, 26

collect(), 58

colwise(), 131

confusion matrix, 208

connected component, 291

connected_components(), 292

convert(), 140, 141

cor(), 158

correctrate(), 209

counts(), 233

CPU_CORES, 326

curve fitting, 269

custom function, 87–91

cutree(), 240

cycle detection, 288–91

DAG. See directed acyclical graph

data cleansing, 104

data discovery, 110–11, 119

data engineering, 102, 103–10

data exploration, 106–8, 118

data frame, 104, 126–35

data learning, 111–13, 119

data modeling, 102, 110–13

data preparation, 103–6, 117

data product, 101

data product creation, 113, 120

data representation, 108–10, 118

data science pipeline, 101–3

data science workflow, 5–7

data type, 49–52

data-driven application, 101

dataset

downloading, 30

loading, 30–32

DBSCAN, 234–37

decision tree, 249–53

deep belief network, 274

deep learning, 112

deep learning system, 274

delete!(), 133

deleterows!(), 133

Density Based Spatial Clustering of Applications with Noise. See DBSCAN

describe(), 129, 157

dictionary, 55–56

DID(), 198

dijkstra_shortest_paths(), 295

Dijktra algorithm, 294, 302

dimensionality reduction, 119, 187

directed acyclical graph, 289

discernibility-based method, 201

discretization, 143

distance calculation function, 34

distance(), 35, 331

distance-based classifier, 33, 142

eachmatch(), 86

ELM. See extreme learning machine

Emac, 18

Epicenter, 17, 18, 20, 45

epoch, 261

Euclidean, 34, 47, 219, 229, 230, 231, 242, 245

extreme learning machine, 265–68

F1 metric, 211

factor analysis, 111

false negative, 171, 172

false positive, 171, 172

feature_type(), 93

Fibonacci, 1

find(), 132, 197

Fisher’s Discriminant Ratio, 149, 151, 194

fit(), 193, 267, 270

fitness function, 200, 201

fitness(), 201

Float32, 51

Float64, 51

FN. See false negative

for-loop, 78

formatting data, 139–41

FP. See false positive

GA. See genetic algorithm approach

Gadfly, 154, 160, 161, 162, 163, 170, 184, 185, 319

generalization, 111

genetic algorithm approach, 200

get(), 69

gini coefficient, 216

GPU. See graphics processing unit

gradient descent, 269

graph

analysis of, 300

dataset for, 285–87

importance of, 282–84

shortest path in a, 294–96

statistics of, 287

graphics processing unit, 275

Graphlab, 41, 75, 127, 134, 329

harmonic mean, 212

has_missing_values(), 93

hcat(), 207

hclust(), 240

HDF5, 41, 46

head(), 129

help, 44

help(), 44

hierarchical clustering, 229, 237–40

histogram, 168–70

hypothesis, 106

hypothesis testing, 170–77

HypothesisTests, 154, 173, 177, 181, 184, 185, 186, 319

IDE. See integrated development environment

if-else statement, 72

IJulia, 16–17

importing and exporting data, 135–37

in, 65

include, 26

index, 52

Index of Discernibility, 148, 149, 151, 194, 197, 203

indmin(), 330

information distillation, 102, 113–16

init_network(), 262

insert!(), 68

insight, deliverance, and visualization, 114, 121

Int32, 50, 51, 57

int32(), 50

Int64, 51, 53, 78, 80, 109, 132, 160

Int8, 51, 52, 61, 140, 147

integrated development environment, 1, 14

intersect(), 176

IPython, 16

ismatch(), 85

isna(), 132, 150

Jaccard Coefficient, 159

Jaccard Similarity, 149, 151, 159, 337

Java, 1, 7, 120, 300, 314

join(), 83

Julia

bridging with Python, 323–24

bridging with R, 321

data science community adoption of, 7–8

interest in, 1

parallelization in, 325–27

Julia Data Format, 41

Julia Studio, 17

JuliaBox, 17, 18, 45, 329

Juno, 14–16

Jupyter, 16, 23, 24, 45

k Nearest Neighbor, 33–39, 45, 274

Kandell’s tau rank correlation, 159

KFCV. See k-fold cross validation

k-fold cross validation, 220–22

K-means, 231–34

kNN. See k Nearest Neighbor

kNN(), 331

kruskal_minimum_spantree(), 298

Kruskal-Wallis test, 172

K-W. See Kruskal-Wallis test

Leave-One-Out, 220, 225

length(), 46, 71, 89, 330

LG2G(), 299, 303

Light Table, 14

line plot, 163

linspace(), 59

listening to the data, 153

load(), 299

logical operator, 76

logistic model, 248

lowercase(), 146

Magic telescope, 27

magic dataset, 27, 32

main(), 94

map(), 74

match(), 85

matchall(), 86

Matlab, 1, 4, 7, 17, 41, 78, 146

matrix, 50

max(), 47, 331

maximal_cliques(), 293

mean, 104

mean square error, 217, 224

mean(), 64, 131

MergeSort, 69

minimum spanning tree, 296–99

misclassification cost, 212

mode(), 94

Monte-Carlo, 325

MSE. See mean square error, See mean square error

MSE(), 248, 272

MST. See minimum spanning tree

Mutual Information, 149, 151, 194, 203, 337

names(), 127

natural language processing, 104, 309

Natural Language Processing, 109

Neobook, 121

nfoldCV_forest(), 259

nfoldCV_tree(), 252

nforldCV_forest(), 257

NLP. See natural language processing

normalization, 142–43, 242

normalize(), 281

notebook

creating, 21

exporting, 24

loading, 24

renaming, 23

saving, 22

Notepad++, 18

nprocs(), 326

nrow(), 134

odd-ratio, 145

OnlineNewsPopularity dataset, 28

Opus Pro, 121

outlier, 137

package

finding and selecting, 18–19

hacking, 21

installing, 20–21

using, 21

pair(), 55

partitional clustering, 229

PCA. See Principal Components Analysis

Pearson’s correlation, 159

Plotly, 161, 184

plots, 160–70

pmap, 327

polynomial, 142

pop!(), 66

precision, 211

predict(), 267

principal component, 188

Principal Components Analysis, 188–93

print(), 56

println(), 57

procs(), 327

Programming Praxis, 1

Project Euler, 1

push!(), 67, 133

Python, 14, 74

quickshift(), 238, 244

quickshiftlabels(), 239

quickshiftplot(), 239

QuickSort, 69

R, 75

RAD. See rapid application development

rand(), 60–63

randn(), 60–63

random forest, 255–60

randperm(), 206

rapid application development, 120

RCE. See Reduced Coulomb Energy

Read, Evaluate, Print, Loop, 13, 16, 51

readlines(), 31

readtable(), 136, 150

recall, 211

receiver operating characteristic curve, 113, 213

Reduced Coulomb Energy, 274

regex, 83–84

regression, 148, See statistical regression

regression tree, 254–55

relative risk transformation, 145

rename!(), 127, 128

rename(), 128

REPL. See Read, Evaluate, Print, Loop

residual variance, 191

ROC curve. See receiver operating characteristic curve

roc(), 215

round(), 59, 267

sample(), 206, 207

sampling, 205–7

SArrays, 41

save(), 42, 299

saving data

delimited file, 40

native Julia format, 41

text file, 43

scatter plot, 164–68

SFrame, 134, 336

SFrames, 41

SGraphs, 41

show(), 58

sig(), 310

signal processing, 109

sil(), 241, 245

silhouette, 240

Silhouette Width, 240

Similarity Index, 151, 159, 337

Simple Matching Coefficient, 159

size(), 71

skewness, 91

skewness(), 157

skewness_type(), 157

sort!(), 69, 133, 134

sort(), 69

Spam Emails dataset, 29

Spearman’s rank correlation, 159

splice!(), 67

split(), 82

sqrt(), 330

SSE. See sum squared error

Stackoverflow, 44, 46

standard deviation, 104

statistical regression, 268–73

StatsBase, 154, 175, 184, 206, 207, 319

stemming, 104

stop word, 104

String, 51

string manipulation, 81–87

string(), 74, 141, 333

subtype, 50, 140

sum squared error, 217, 218

sum(), 63

summarystats(), 157

supervised machine learning, 247–48

support, 44

support vector machine, 273

SVM. See support vector machine

symbol(), 128

tail(), 129

time(), 71

total misclassification cost, 213

train(), 262, 263

Transductive Support Vector Machine, 274

transductive system, 274

transform(), 192, 193

true negative, 172

true positive, 172

t-SNE, 166, 182

TSVM. See Transductive Support Vector Machine

t-test, 173

Tutorials Point, 18

typemax(), 57

typemin(), 57

typeof(), 50

unsupervised learning, 228–31

uppercase(), 146

validation, 112

var(), 131

Variation of Information, 240, 241, 245

vector, 50, 60

vector2ANN(), 262

vectorization, 146

Vega, 161, 184

weighted accuracy, 209

while-loop, 79

Winston, 161, 184

working directory, 26

wrapper function, 25, 35, 37, 46, 325, 326, 330

writecsv(), 40

writedlm(), 40

writetable(), 136, 150

zeros(), 196