资源描述:
《Mining_of_Massive_Datasets 大数据-互联网大规模数据挖掘与分布式处理.pdf》由会员上传分享,免费在线阅读,更多相关内容在教育资源-天天文库。
1、MiningofMassiveDatasetsAnandRajaramanKosmix,Inc.JeffreyD.UllmanStanfordUniv.Copyrightc2010AnandRajaramanandJeffreyD.UllmaniiPrefaceThisbookevolvedfrommaterialdevelopedoverseveralyearsbyAnandRaja-ramanandJeffUllmanforaone-quartercourseatStanford.ThecourseCS345A,t
2、itled“WebMining,”wasdesignedasanadvancedgraduatecourse,althoughithasbecomeaccessibleandinterestingtoadvancedundergraduates.WhattheBookIsAboutAtthehighestlevelofdescription,thisbookisaboutdatamining.However,itfocusesondataminingofverylargeamountsofdata,thatis,
3、datasolargeitdoesnotfitinmainmemory.Becauseoftheemphasisonsize,manyofourexamplesareabouttheWebordataderivedfromtheWeb.Further,thebooktakesanalgorithmicpointofview:dataminingisaboutapplyingalgorithmstodata,ratherthanusingdatato“train”amachine-learningengineofso
4、mesort.Theprincipaltopicscoveredare:1.Distributedfilesystemsandmap-reduceasatoolforcreatingparallelalgorithmsthatsucceedonverylargeamountsofdata.2.Similaritysearch,includingthekeytechniquesofminhashingandlocality-sensitivehashing.3.Data-streamprocessingandspec
5、ializedalgorithmsfordealingwithdatathatarrivessofastitmustbeprocessedimmediatelyorlost.4.Thetechnologyofsearchengines,includingGoogle’sPageRank,link-spamdetection,andthehubs-and-authoritiesapproach.5.Frequent-itemsetmining,includingassociationrules,market-bas
6、kets,theA-PrioriAlgorithmanditsimprovements.6.Algorithmsforclusteringverylarge,high-dimensionaldatasets.7.TwokeyproblemsforWebapplications:managingadvertisingandrec-ommendationsystems.iiiivPREFACEPrerequisitesCS345A,althoughitsnumberindicatesanadvancedgraduat
7、ecourse,hasbeenfoundaccessiblebyadvancedundergraduatesandbeginningmastersstudents.Inthefuture,itislikelythatthecoursewillbegivenamezzanine-levelnumber.TheprerequisitesforCS345Aare:1.Thefirstcourseindatabasesystems,coveringapplicationprogramminginSQLandotherdat
8、abase-relatedlanguagessuchasXQuery.2.Asophomore-levelcourseindatastructures,algorithms,anddiscretemath.3.Asophomore-levelcourseinsoftwaresystems,softwareengineering,andprogramminglanguage