集体智慧编程

Foreword

Preface

1.IntroductiontoCollectiveIntelligence

WhatIsCollectiveIntelligence?

WhatIsMachineLearning?

LimitsofMachineLearning

Real-LifeExamples

OtherUsesforLearningAlgorithms

2.MakingRecommendations

CollaborativeFiltering

CollectingPreferences

FindingSimilarUsers

RecommendingItems

MatchingProducts

Buildingadel.icio.usLinkRecommender

Item-BasedFiltering

UsingtheMovieLensDataset

User-BasedorItem-BasedFiltering?

Exercises

3.DiscoveringGroups

SupervisedversusUnsupervisedLearning

WordVectors

HierarchicalClustering

DrawingtheDendrogram

ColumnClustering

K-MeansClustering

ClustersofPreferences

ViewingDatainTwoDimensions

OtherThingstoCluster

Exercises

4.SearchingandRanking

WhatsinaSearchEngine?

ASimpleCrawler

BuildingtheIndex

Querying

Content-BasedRanking,

UsingInboundLinks

LearningfromClicks

Exercises

5.Optimization

GroupTravel

RepresentingSolutions

TheCostFunction

RandomSearching

HillClimbing

SimulatedAnnealing

GeneticAlgorithms

RealFlightSearches

OptimizingforPreferences

NetworkVisualization

OtherPossibilities

Exercises

6.DocumentFiltering

FilteringSpam

DocumentsandWords

TrainingtheClassifier

CalculatingProbabilities

ANaiveClassifier

TheFisherMethod

PersistingtheTrainedClassifiers

FilteringBlogFeeds

ImprovingFeatureDetection

UsingAkismet

AlternativeMethods

Exercises

7.ModelingwithDecisionTrees

PredictingSignups

IntroducingDecisionTrees

TrainingtheTree

ChoosingtheBestSplit

RecursiveTreeBuilding

DisplayingtheTree

ClassifyingNewObservations

PruningtheTree

DealingwithMissingData

DealingwithNumericalOutcomes

ModelingHomePrices

Modeling"Hotness"

WhentoUseDecisionTrees

Exercises

8.BuildingPriceModels

BuildingaSampleDataset

k-NearestNeighbors

WeightedNeighbors

Cross-Validation

HeterogeneousVariables

OptimizingtheScale

UnevenDistributions

UsingRealData--theeBayAPI

WhentoUsek-NearestNeighbors

Exercises

9.AdvancedClassification:KernelMethodsandSVMs

MatchmakerDataset

DifficultieswiththeData

BasicLinearClassification

CateRoricalFeatures

ScalingtheData

UnderstandingKernelMethods

Support-VectorMachines

UsingLIBSVM

MatchingonFacebook

Exercises

10.FindingIndependentFeatures

ACorpusofNews

PreviousApproaches

Non-NegativeMatrixFactorization

DisplayingtheResults

UsingStockMarketData

Exercises

11.EvolvingIntelligence

WhatIsGeneticProgramming?

ProgramsAsTrees

CreatingtheInitialPopulation

TestingaSolution

MutatingPrograms

Crossover

BuildingtheEnvironment

ASimpleGame

FurtherPossibilities

Exercises

12.AlgorithmSummary

BayesianClassifier

DecisionTreeClassifier

NeuralNetworks

Support-VectorMachines

k-NearestNeighbors

Clustering

MultidimensionalScaling

Non-NegativeMatrixFactorization

Optimization

A.Third-PartyLibraries

B.MathematicaIFormulas

Index

　　想要探寻搜索排名、产品推荐、社会化书签和在线匹配背后的力量吗？这本颇具魅力的书籍向你展现如何创建Web2.0应用程序，从参与性Internet应用程序产生的大量数据中挖掘金矿。运用本书中介绍的先进算法，你可以编写聪明的程序，以访问其他网站那些有趣的数据集，从自有应用程序的用户中收集数据，或者分析和理解你所发现的数据。　　《集体智慧编程》将你带入机器学习和统计的世界，并且阐释了如何从你和他人每天收集的信息中获得关于用户体验、市场营销、个性品味及人类行为的结论。每个算法的描述都十分简明清晰，相关代码均可以立即用于你的网站、博客、Wiki或特定应用程序。本书讲解了下列主题：　　可以让在线零售商推荐产品或媒体的协作过滤技术；用于在大数据集中发现同类项组的聚类方法；从数以百万计可能方案中选择问题最佳解决方案的最优化算法；贝叶斯过滤，用在基于单词类型和其他特征的垃圾信息过滤中；支持向量（support-vector）机器，用于在线交友网站中的速配；用于问题解决的演化智能计算机如何通过多次玩同样的游戏，改进自身代码并获得技能提升。　　每一章都包含了相关练习，可通过扩展使算法变得更强大。超越简单的数据库支持应用程序模式，让Internet数据财富为你所用。