• Choose a category
• All
• Classic-Fiction

holdout method370, 386

random sampling370, 386

recursive335

tuples334

Partitioning Around Medoids (PAM) algorithm455–457

partitioning methods448, 451–457, 491

centroid-based451–454

global optimality449

iterative relocation techniques448

k-means451–454

k-medoids454–457

k-modes454

object-based454–457 see alsocluster analysis

path-based similarity594

pattern analysis, in recommender systems282

pattern clustering308–310

pattern constraints297–300

pattern discovery601

pattern evaluation8

pattern evaluation measures267–271

all_confidence268

comparison269–270

cosine268

Kulczynski268

max_confidence268

null-invariant270–271 see alsomeasures

pattern space pruning295

pattern-based classification282, 318

pattern-based clustering282, 516

Pattern-Fusion302–307

characteristics304

core pattern304–305

initial pool306

iterative306

merging subpatterns306

shortcuts identification304 see alsocolossal patterns

pattern-guided mining30

patterns

actionable22

co-location319

colossal301–307, 320

combined significance312

constraint-based generation296–301

context modeling of314–315

core304–305

distance309

evaluation methods264–271

expected22

expressed309

frequent17

hidden meaning of314

interesting21–23, 33

metric space306–307

negative280, 291–294, 320

negatively correlated292, 293

rare280, 291–294, 320

redundancy between312

relative significance312

representative309

search space303

strongly negatively correlated292

structural282

type specification15–23

unexpected22 see alsofrequent patterns

pattern-trees264

Pearson' correlation coefficient222

percentiles48

perception-based classification (PBC)348

illustrated349

as interactive visual approach607

pixel-oriented approach348–349

split screen349

tree comparison350

phylogenetic trees590

pivot (rotate) operation148

pixel-oriented visualization57

planning and analysis tools153

point queries216, 217, 220

pool-based approach433

positive correlation55, 56

positive tuples364

positively skewed data47

possibility theory428

posterior probability351

postpruning344–345, 346

power law distribution592

precision measure368–369

predicate sets

frequent288–289

k289

predicates

repeated288

variables295

prediction19

classification328

link593–594

loan payment608–609

with naive Bayesian classification353–355

numeric328, 385

prediction cubes227–230, 235

example228–229

Probability-Based Ensemble229–230

predictive analysis18–19

predictive mining tasks15

predictive statistics24

predictors328

prepruning344, 346

prime relations

contrasting classes175, 177

deriving174

target classes175, 177

principle components analysis (PCA)100, 102–103

application of103

correlation-based clustering with511

illustrated103

in lower-dimensional space extraction578

procedure102–103

prior probability351

privacy-preserving data mining33, 621, 626

distributed622

k-anonymity method621–622

l-diversity method622

as mining trend624–625

randomization methods621

results effectiveness, downgrading622

probabilistic clusters502–503

probabilistic hierarchical clustering467–470

agglomerative clustering framework467, 469

algorithm470

drawbacks of using469–470

generative model467–469

interpretability469

understanding469 see alsohierarchical methods

probabilistic model-based clustering497–508, 538

expectation-maximization algorithm505–508

fuzzy clusters and499–501

product reviews example498

user search intent example498 see alsocluster analysis

probability

estimation techniques355

posterior351

prior351

probability and statistical theory601

Probability-Based Ensemble (PBE)229–230

PROCLUS511

profiles614

proximity measures67

for binary attributes70–72

for nominal attributes68–70

for ordinal attributes74–75

proximity-based methods552, 560–567, 581

density-based564–567

distance-based561–562

effectiveness552

example552