Data Smart: Using Data Science to Transform Information into Insight

Index

A

absolute references, Solver, 110

absolute values of errors, median regression, 221

additive smoothing, 86

adjacency matrix, 158

affinity matrix, 159

agglomerative clustering, 185

AI model

Bayes rule and, 83–86

dummy variables, 210–212

feature set, 207–208

versus optimization model, 101–102

overview, 206–207

pregnancy data, 378–385

pregnant customers (See RetailMart (pregnant customers))

training data, oversampling, 210

AIMMS, 118

algorithms, evolutionary, 115–116

alpha value calculation, 276–277

arrays, formulas, 19–20

autocorrelations, 306–313

B

bag of words model, 79

extraneous punctuation, 87–88

spaces, 88–91

bagged decision stumps, 251

bagging, 254. See also decision stumps

model evaluation, 267–271

outliers and, 271

random forest models, 271

Bayes rule, 82

AI model creation, 83–86

Big M,133–137

binary tree, 193–197

BINOMDIST function, 116

blending model, 119

boosting, 251

model evaluation, 280–283

model training, 272–275

weighted errors, 272

reweighting, 277–278

C

CDF (cumulative distribution function), 146–148, 337

mean deviation, 147–148

standard deviation, 147–148

scenarios from, 148–150

cell formatting, 5–7

central limit theorem, 146

chain rule of probability, 81

charts. See also graphs

fan chart, 331–333

inserting, in spreadsheets, 8–9

classifiers, bagging, 254

cluster analysis, 29

cluster centers, solving for, 46–48

cluster labels, 193–197

clustering, 29–30

agglomerative, 185

cluster centroid, 31

community detection, 155–156

divisive, 185–192

hierarchical partitioning, ...