• Choose a category
• All
• Classic-Fiction

average()215

B

background knowledge30–31

backpropagation393, 398–408, 437

activation function402

algorithm illustration401

biases402, 404

case updating404

efficiency404

epoch updating404

error403

functioning of400–403

hidden layers399

input layers399

input propagation401–402

interpretability and406–408

learning400

learning rate403–404

logistic function402

multilayer feed-forward neural network398–399

network pruning406–407

neural network topology definition400

output layers399

sample learning calculations404–406

sensitivity analysis408

sigmoid function402

squashing function403

terminating conditions404

unknown tuple classification406

weights initialization401 see alsoclassification

bagging379–380

algorithm illustration380

boosting versus381–382

in building random forests383

bar charts54

base cells189

base cuboids111, 137–138, 158

Basic Local Alignment Search Tool (BLAST)591

Baum-Welch algorithm591

Bayes’ theorem350–351

Bayesian belief networks393–397, 436

algorithms396

components of394

conditional probability table (CPT)394, 395

directed acyclic graph394–395

gradient descent strategy396–397

illustrated394

mechanisms394–396

problem modeling395–396

topology396

training396–397 see alsoclassification

Bayesian classification

basis350

Bayes’ theorem350–351

class conditional independence350

naive351–355, 385

posterior probability351

prior probability351

BCubed precision metric488, 489

BCubed recall metric489

behavioral attributes546, 573

believability, data85

BI (business intelligence)27

biases402, 404

biclustering512–519, 538

application examples512–515

enumeration methods517, 518–519

gene expression example513–514

methods517–518

optimization-based methods517–518

recommender system example514–515

types of538

biclusters511

with coherent values516

with coherent values on rows516

with constant values515

with constant values on columns515

with constant values on rows515

as submatrix515

types of515–516

bimodal47

bin boundaries89

binary attributes41, 79

asymmetric42, 70

as Boolean41

contingency table for70

dissimilarity between71–72

example41–42

proximity measures70–72

symmetric42, 70–71 see alsoattributes

binning

discretization by115

equal-frequency89

smoothing by bin boundaries89

smoothing by bin means89

smoothing by bin medians89

biological sequences586, 624

alignment of590–591

analysis590

BLAST590

hidden Markov model591

as mining trend624

multiple sequence alignment590

pairwise alignment590

phylogenetic tree590

substitution matrices590

bipartite graphs523

BIRCH458, 462–466

CF-trees462–463, 464, 465–466

clustering feature462, 463, 464

effectiveness465

multiphase clustering technique464–465 see alsohierarchical methods

bitmap indexing160–161, 179

bitmapped join indexing163, 179

bivariate distribution40

BLAST. seeBasic Local Alignment Search Tool

BOAT.

Boolean association rules281

Boolean attributes41

boosting380

accuracy382

AdaBoost380–382

bagging versus381–382

weight assignment381

bootstrap method371, 386

bottom-up design approach133, 151–152

bottom-up subspace search510–511

boxplots49

computation50

example50

five-number summary49

illustrated50

in outlier visualization555

BUC200–204, 235

for 3-D data cube computation200

algorithm202

Apriori property201

bottom-up construction201

iceberg cube construction201

partitioning snapshot203

performance204

top-down processing order200, 201

business intelligence (BI)27

business metadata135

business query view151

C

C4.5332, 385

class-based ordering358

gain ratio use340

greedy approach332

pessimistic pruning345

rule extraction358 see alsodecision tree induction

cannot-link constraints533

CART332, 385

cost complexity pruning algorithm345

Gini index use341

greedy approach332 see alsodecision tree induction

case updating404

case-based