Machine Learning Resource Guide
User Manual:
Open the PDF directly: View PDF .
Page Count: 15
MachineLearningMastery
Web: http://MachineLearningMastery.com
Email: jason@MachineLearningMastery.com
MachineLearningResourceGuide
byJasonBrownlee,PhD
Copyright©2014JasonBrownlee,AllRightsReserved.
SharethisGuide
Ifyouknowsomeonewhocanbenefitfromthisguide,justsendthemthislink:
http://MachineLearningMastery.com/machinelearningresources/
2/15 http://MachineLearningMastery.com
Table of Contents
Introduction
Books
BeginnerBooks
PracticalBooks
PythonBooks
RBooks
Textbooks
Communities
StackExchange
Reddit
Quora
Other
Videos
UniversityCourses
PaidCourses
OtherVideos
UniversityCourseMaterial
UndergraduateLevel
GraduateLevel
SoftwareandLibraries
Competitions
Guides
Beginner
Novice
Intermediate
JetFuel
ConnectWithMe!
3/15 http://MachineLearningMastery.com
Introduction
Hithere,mynameisJasonfromMachineLearningMastery.com.Thanksfordownloadingthis
MachineLearningResourceGuide.
Ihaveworkedhardtocollectandlistonlythebestresourcesthatwillhelpyoujumpstartyour
journeytowardsmachinelearningmastery.I’vecategorizedtheresourcesintomainthemes
suchasvideos,booksandcourses.
I’mcertainyouwillfindgreatvalueintheresourceslistedinthisguide.Takeyourtimeandselect
amediumorresourcetypeyoupreferandstartworkingthroughresourcesonebyone.Trynot
tooverloadyourself.Remembertothinkhardaboutwhatyouwantfromaresourceandactively
takenotes.
I’minterestedtohearwhatresourcesyoutry,sendmeanemailandletmeknowvia
jason@MachineLearningMastery.comorvisitmysiteMachineLearningMastery.comandleavea
comment.Ihopetohearfromyousoon.
JasonBrownlee.
4/15 http://MachineLearningMastery.com
Books
Ireadalotofbooks,andeveninthisageofebooks,Ilikehavingalotofreferencebooksonthe
bookshelf.IalsolikehavingbooksinPDFformsoIcansearchthemquicklyandpulloutthe
informationIneed.
Thebookslistedinthissectionaregroupedbyafewdifferentcriteriathatyoumayfinduseful,as
suchyoumayseeafewduplicatesacrossthedifferentlistsofbooks.
IhaveprovidedlinkstoeachbookonAmazon.ThelinksareaffiliatelinkswhichmeansIwillget
afewcentsfromAmazonifyoudecidetobuyabook.
Myadvice:Pickonebookandreadit,covertocover.
Beginner Books
Thesearebooksfortheabsolutebeginnertogetafeelingforwhatmachinelearningorworking
withdataisallabout.Fromabusinessandsemitechnicalperspective.
●PredictiveAnalytics:ThePowertoPredictWhoWillClick,Buy,Lie,orDie
●DataScienceforBusiness:Whatyouneedtoknowaboutdatamininganddataanalytic
thinking
●DataSmart:UsingDataSciencetoTransformInformationintoInsight
Practical Books
Ifyouareaprogrammerorengineerandarelookingforabookwithcodeexamplestoimplement
orexecute,thesearebooksforyou:
●ProgrammingCollectiveIntelligence:BuildingSmartWeb2.0Applications
●DataMining:PracticalMachineLearningToolsandTechniques
●MachineLearningforHackers
●MachineLearning:AnAlgorithmicPerspective
●MachineLearninginAction
●AppliedPredictiveModeling
Youcanlearnmoreaboutthesebooksinmyblogpost6PracticalBooksforBeginningMachine
Learning
Python Books
Thesearebooksforlearningandapplyingmachinelearningifyouareapythonprogrammer.
●BuildingMachineLearningSystemswithPython
●Learningscikitlearn:MachineLearninginPython
●MachineLearninginAction
5/15 http://MachineLearningMastery.com
●ProgrammingCollectiveIntelligence:BuildingSmartWeb2.0Applications
●MachineLearning:AnAlgorithmicPerspective
●MiningtheSocialWeb:DataMiningFacebook,Twitter,LinkedIn,Google+,GitHub,and
More
●NaturalLanguageProcessingwithPython
●ProgrammingComputerVisionwithPython:Toolsandalgorithmsforanalyzingimages
●PythonforDataAnalysis:DataWranglingwithPandas,NumPy,andIPython
Youcanlearnmoreaboutthesebooksinmyblogpost:PythonMachineLearningBooks
R Books
IfyouareanRprogrammerorarelookingatapplyingmachinelearninginR,thesebooksarefor
you.
●AppliedPredictiveModeling
●AnIntroductiontoStatisticalLearning:withApplicationsinR
●PracticalDataSciencewithR
●MachineLearningwithR
●DataMiningwithR:LearningwithCaseStudies
●DataMiningandBusinessAnalyticswithR
●DataMiningwithRattleandR:TheArtofExcavatingDataforKnowledgeDiscovery
Youcanlearnmoreaboutthesebooksinmyblogpost:BooksforMachineLearningwithR
Textbooks
Thesearebooksformachinelearningpractitionerslookingtogobeyondthepracticalbooksand
deeperintotheory.Thesearetextbookscommonlyusedinundergraduateandpostgraduate
universitycourses.
●MachineLearning,byTomMitchell
●LearningFromData,byYaserAbuMostafa,MalikMagdonIsmailandHsuanTienLin
●MachineLearning:AProbabilisticPerspective,byKevinMurphy
●PatternRecognitionandMachineLearning,byChristopherBishop
●TheElementsofStatisticalLearning:DataMining,Inference,andPrediction,byTrevor
Hastie,RobertTibshiraniandJeromeFriedman
6/15 http://MachineLearningMastery.com
Communities
Youwillhavealotofquestionsalongyourjourneytowardmachinelearningmasteryandthere
areexcellentplaceswheremachinelearningexpertscananswerthosequestionsforyou,ifyou
knowwheretolook.
Eachsitelistedbelowallowsyoutocreateanaccountforfreeandaskyourquestion.Review
thetypesofquestionsandanswersofferedineachcommunitybeforeselectingtheright
communityforyoutoaskyourquestion.
Itisverylikelyyourquestionhasbeenaskedandansweredbefore.Trysearchingforitoneach
communitybeforeposting.
Stack Exchange
Thestackexchangesitesarequestionandanswercommunities,sotheyaretargetedtowards
problemsolving.Youcanpostthespecificquestionsyouhave,answerquestionstowhichyou
knowtheanswerand(myfavorite)readquestionsandanswerstodiscovernewmethodsand
perspectives.
TherearefoursitesIliketodipinto:
●CrossValidated:Thissiteisusefulforlowlevelquestionsonalgorithmsandstatistical
methods.
●QuantitativeFinance:(specificallythemachinelearningtag)Thissiteisusefulifyouare
operatinginthefinancialdomain,butgenerallyifyouareworkingwithtimeseriesdata.
●Programmers:(specificallythemachinelearningtag)Greatforspecificcodequestions,
suchasaproblemwithagivenlibraryortoolyouareusing.
●StackOverflow:(specificallythemachinelearningtag)Again,likeprogrammers,greatfor
specificquestionswiththeimplementationsideofmachinelearning.It’salsotheoldest
siteandcancovermachinelearningalgorithmsandlibraries.
Thereisanewsitethathasstartedup,butisstillinbeta,soitmaynotsurvive.ItiscalledData
ScienceandIamfindingitveryinterestingforthegeneralconcernsofappliedmachinelearning
(mixofcodeandmath).
Reddit
Redditisacommunityofcommunitiescalledsubreddits.Agivensubredditcanbequestionand
answersite,alinksharingsiteor(moretypically)amixofthetwo.
AfewsubredditsIfrequentinclude:
●MachineLearning:Containsofmixof“howdoIgetstarted”andmoreadvancedlinksto
machinelearningblogposts.Alsogoodforlinkingtoyourownprojectstogetsome
feedback.
7/15 http://MachineLearningMastery.com
●ComputerVision:Mostlyquestionsoncomputervisionquestionsboththeoreticaland
practical(suchaslibraries).
●NaturalLanguage:Focusonnaturallanguageprocessing,providingagoodmixof
questionsandlinkstorelevantarticlesandblogposts.
●Statistics:Discussiononstatisticalsoftwareandmethods,greatfordiggingdeeperintoa
givenmethodoralgorithm.
●DataScience:Mostlylinkstopoststhatstraddledataanalysisandmachinelearning.
●BigData:Focusedpostsanddiscussionsonthebigdataecosystem.
Thereareothersubredditsonrelevantandrelatedtopics,butIhavenotfoundthemasuseful.
Quora
Quoraisaquestionandanswersitethatisdividedintotopics,muchlikeredditbutonly
questionsandanswers.Thequestionsaretypicallygoodandtheanswershighquality.Unlike
thestackexchangesites,theyaretypicallylesstechnical,lessproblemfocusedandmore
meaty.
AfewQuoratopicsIfrequentinclude:
●MachineLearning:Usefulforhighlevelquestionsonalgorithms,processes,resources
andgettingstarted.Agoodmix.
●Statistics:Focusondeeperstatisticalmethodsandalgorithms,butincludesalotof
machinelearningcontent.
●DataMining:Goodquestionswithafocusontheappliedsideofmachinelearning,buta
lotofoverlapwithMachineLearning.
●DataScience:MuchliketheDataMiningandMachinelearningtopics,thequestionsare
typicallyahigherlevel.
Therearemanyothertopicsthatmightbeuseful,notlimitedtoDataAnalysis,Predictive
Analytics,NLPandComputerVision.AlsotherearetopicsonspecificmethodssuchasSVM,
DeepLearning,Classification,andR.
Other
TherearesomeothergreatcommunitiesaroundthatIcouldnotclassifyaseasily.
●MetaOptimizeQ+A:LikeCrossValidated,thisisaquestionandanswersitethatisgreat
forlowerlevelquestionsonspecificalgorithmsandmethods.Mathsandtheoryheavy.
●KaggleForums:Greatfordiscussionaroundspecificcompetitionsanddatasets,andfull
ofgreatnuggetsofadviceforfeatureengineering,ensemblingandrefiningyourtest
harnesses.
●DataTau:Asocialnewssitewithafocusonlinkstopostsondataandmachinelearning
relatetopics.Lowtrafficandusefullinks.
8/15 http://MachineLearningMastery.com
Videos
Videosareagreatwaytolearnaboutmachinelearning,bothforlectureandtutorialcontent.
University Courses
ThereareuniversitycoursesthatareofferedonlineforfreebyorganizationssuchasCoursera
andedX.Theyincludevideolectures,homeworkthatisassessed,quizzesandtests.Someof
thecoursesalsohavejustthevideolectureslistedonsiteslikeYouTube.Trysearching.
●Stanford:MachineLearning,byAndrewNg
●Stanford:ProbabilisticGraphicalModels,byDaphneKoller
●Caltech:LearningfromData,byYaserAbuMostafa
●UniversityofToronto:NeuralNetworksforMachineLearning,byGeoffreyHinton
●UniversityofWashington:MachineLearning,byPedroDomingos
●UniversityofWashington:IntroductiontoRecommenderSystems,byJosephKonstan
andMichaelEkstrand
●UniversityofWashington:IntroductiontoDataScience,byBillHowe
Paid Courses
TherearepaidcoursesonMachineLearningofferedbyorganizationssuchasUdemy.Youpay
afeeandhaveaccesstothepremiumcontenttolearnsomethingspecific.
●Udemy:AnIntrotoMachineLearningwithWebData,byHilaryMason
●Udemy:AdvancedMachineLearning,byHilaryMason
●Udemy:IntroductiontoR,byJagannathRajagopal
●Udemy:WorkingwithBigData,byPearson
Other Videos
●MachineLearningCategoryonVideoLectures.Net
●“GettingInShapeForTheSportOfDataScience”TalkbyJeremyHoward
●FacebookTechTalk:PeterNorvigonbigdata
9/15 http://MachineLearningMastery.com
University Course Material
Thereisapopulartrendfortopleveltechnicaluniversitiestoputcoursematerialsonline
includinglecturevideos,slides,homeworkandassignments.Thismaterialcanbeusedfor
selfstudy.
Someuniversitiesmakethematerialseasiertofindthanothers,MITisashininglightinthis
regardwiththeirOpenCourseWareinitiative.
Undergraduate Level
●MIT6.034ArtificialIntelligence(provideamachinelearningfocus)
●MIT15.075StatisticalThinkingandDataAnalysis
●StanfordCS229MachineLearning(SEEsite)
●StanfordStatistics315aModernAppliedStatistics:ElementsofStatisticalLearning
●StanfordStatistics315bModernAppliedStatistics:ElementsofStatisticalLearningII
●CaltechLearningfromData
Graduate Level
●MIT6.867MachineLearning
●MIT6.825TechniquesinArtificialIntelligence(relatedmachinelearningtopics)
●MIT9.520StatisticalLearningTheoryandApplications
●MIT9.641IntroductiontoNeuralNetworks
●MIT15.097Prediction:MachineLearningandStatistics
●MIT18.465TopicsinStatistics:StatisticalLearningTheory
●HarvardCS281IntelligentMachines:Perception,Learning,andUncertainty(alsoCS181)
●CornellCS6784AdvancedMachineLearning
●CMU10701MachineLearning(videoshereandhere)
10/15 http://MachineLearningMastery.com
Software and Libraries
Therearealotofsoftwareandlibrariesthatyoucoulduseformachinelearning.
Belowaresomebestofbreedsoftwaretoolsandlibrariesthatareusefulforlearningand
practicingmachinelearning.
●WEKA(GUI,Java)
●R
●ScikitLearn(Python)
●Octave(anopensourceMatLab)
●BigML(inthebrowser)
Ifyouarejuststartingout,IrecommendusingtheWekagraphicaluserinterface.Forexample,
youcanrunyourfirstclassifierin5minutesflat.
Ifyouarestrugglingwithwhichprogramminglanguagetouse,checkoutmypost:
BestProgrammingLanguageforMachineLearning.
IfyouareaJavaprogrammeryoumaybeinterestedinmypost:JavaMachineLearning.
11/15 http://MachineLearningMastery.com
Competitions
CompetitionsarecommonwithArtificialIntelligenceandMachineLearningconferences.Takea
lookatthewebpagesforsomeofthepopularconferencesandyouwillverylikelyfindcurrent
activemachinelearningcompetitions.
Competitivemachinelearningcanbeagreatwaytolearnnewdatapreparationandmodelling
techniques.Peopleinandaroundcompetitivemachinelearningcanprovideawealthoftips,
resourcesanddifferentwaysofapproachingthesameproblem.Thecompetitionscanalsobea
greatwaytotestoutmethodsandideas.
●Kaggle
●TunedIT
●CrowdAnalytix
●InnoCentive
●Challenge.gov
●KDDCup
12/15 http://MachineLearningMastery.com
Guides
Ihaveapassionforhelpingprogrammersandengineersgetstartedandkickasswithmachine
learning.Youcanshortcutyourmachinelearningjourneywithanumberoftheguidesand
coursesthatIhavecreatedforyou.
Beginner
●SelfStudyGuidetoMachineLearning:(StartHere!)Discoverthestructuredframework
forselfstudyingmachinelearningthatincludes4competencylevelsandfocused
objectivesandactivitiesforeachlevel.
●MachineLearningFoundations:Discovertheconceptsanddefinitionsofmachine
learningandhavetheconfidencetoexplainittofriendsandcolleagues.
●ConquerSelfLimitingBeliefsinMachineLearning:Discoveryourownselflimitingbeliefs
thatarehaltingyoufromgettingstartedormakingprogressinthefieldofmachine
learning.
●MachineLearningMatters:Discoverwhymachinelearningmatterstoyouandwhyit
matterstotheworld.
Novice
●AppliedMachineLearningProcess:Discoverthestructuredstepbystepprocessfor
applyingmachinelearningtoyourownproblemsnowandinthefuture.Includesaclear
6stepframeworkwithactivitiesandquestionstoanswerateachstepalongtheway.
●JumpStartScikitLearn:DiscoverthePythonmachinelearninglibraryscikitlearninthis
lightweightrecipebook.Contains35recipesreadytocopyandpastefordatahandling,
supervisedlearning,regularizationalgorithms,ensemblemethodsandadvancedtopics.
●JumpStartWeka:DiscovertheWekamachinelearningworkbenchincluding
stepbysteptutorialsforanalyzingdata,applyingmachinelearningalgorithmsand
designingandinterpretingmachinelearningexperiments.
●BeginningWeka:[VideoCourse]Discovertheprocessofappliedmachinelearningwith
stepbysteptutorialsandworkedcasestudyproblemsusingtheWekamachinelearning
workbench.Thefeatureofthiscoursearethe3realworldcasestudieswithstepbystep
tutorialsandvideos.
Intermediate
●SmallProjectsMethodology:Discovertheblueprintforlearningandpracticingapplied
machinelearningwith4projecttypesand90projectideas.
●AlgorithmDescriptionTemplate:Discoverastrategyforlearningamachinelearning
algorithmfast.Iusedthisstrategytolearnanddescribe45natureinspiredalgorithms
thatIturnedintoabook.
●CleverAlgorithms:NatureInspiredProgrammingRecipes:Discover45natureinspired
algorithmsdescribedconsistentlyusingastructuredalgorithmtemplate.Allalgorithms
includeaworkingimplementationinRuby.
13/15 http://MachineLearningMastery.com
Jet Fuel
●SuperBundle:[GetItAll!]Inthisbundleyougetacopyofallcurrentandallfuture
standalonemachinelearningmasteryguidesandcourses.Asapartofthissuperbundle
youwillbeemailedasnewguidesareaddedinthefuturesothatyoucandownloadthem
atnoextracost.
14/15 http://MachineLearningMastery.com
Connect With Me!
Hey,mynameisJason.I’mfromAustraliaandIhaveaMastersandPhdinArtificialIntelligence,
I’vewrittenbooksonalgorithms,consultedforstartupsandIworkontropicalcycloneforecasting
systems.Igetalotofsatisfactionhelpingprogrammersmaketheirstartandkicksomeasswith
machinelearning.
Iam33yearsold,marriedwithayoungsonandinmyfreetimeIliketoreadbooks,code,write
articlesandparticipateinmachinelearningcompetitions.
Youcanlearnmoreaboutmeandmystorybyclickinghere.
Reachouttome,I’dlovetohearfromyouandyourgoalswithmachinelearning.
Contactmeviaemailonjason@MachineLearningMastery.com
Followmeon:
LinkedIn: https://www.linkedin.com/in/jasonbrownlee
Twitter: http://twitter.com/TeachTheMachine
Facebook: http://www.facebook.com/pages/MachineLearningMastery/1429846323896563
Google+: http://plus.google.com/+MachineLearningMasteryHome
15/15 http://MachineLearningMastery.com