how computers know what we want - before we do

how computers know what we want - before we do

ID:13575327

大小:21.64 KB

页数:10页

时间:2018-07-23

how computers know what we want - before we do_第1页
how computers know what we want - before we do_第2页
how computers know what we want - before we do_第3页
how computers know what we want - before we do_第4页
how computers know what we want - before we do_第5页
资源描述:

《how computers know what we want - before we do》由会员上传分享,免费在线阅读,更多相关内容在行业资料-天天文库

1、HowComputersKnowWhatWeWant-BeforeWeDoHere’sanexperiment:trythinkingofasongnotasasongbutasacollectionofdistinctmusicalattributes.Maybethesonghaspoliticallyrics.Thatwouldbeanattribute.Maybeithasapolicesireninit,oraprominentbanjopart,orpairedvocalharmony,orpunkroots.Anyoneofthosew

2、ouldbeanattribute.Asongcanhaveasmanyas400attributes-thosearejustafewoftheonesfiledunderp.ThiscuriousideaoriginatedwithTimWestergren,oneofthefoundersofanInternetradioservicebasedinOakland,Calif.,calledPandora.Everytimeanewsongcomesout,someoneonPandora’sstaff-aspeciallytrainedmus

3、icianormusicologist-goesthroughalistofpossibleattributesandassignsthesonganumericalratingforeachone.Analyzingasongtakesabout20minutes.ThepeopleatPandora-norelationtothealienplanet-analyze10,000songsamonth.They’vebeendoingitfor10yearsnow,andsofarthey’veamassedadatabasecontaining

4、detailedprofilesof740,000differentsongs.WestergrencallsthisdatabasetheMusicGenomeProject.Thereisapointtoallthis,apartfromsettlingbarbetsaboutwhichsonghasthemostprominentbanjopartever.ThepurposeoftheMusicGenomeProjectistomakepredictionsaboutwhatkindofmusicyou’regoingtolikenext.P

5、andorausestheMusicGenomeProjecttopowerwhat’sknowninthebusinessasarecommendationengine:oneofthosepiecesofsoftwarethatgivesyouadviceaboutwhatyoumightenjoylisteningtoorwatchingorreadingnext,basedonwhatyoujustlistenedtoorwatchedorread.TellPandorayoulikeSpoonandit’llplayyouModestMou

6、se.TellityoulikeCajunaccordionvirtuosoAlphonse“BoisSec”Ardoinandit’lltryyououtonsomeIryLeJeune.EnoughpeopleliketellingPandorawhattheylikethattheserviceadds2.5millionnewusersamonth.Overthepastdecade,recommendationengineshavebecomequietlyubiquitous.Attheappropriatemoment-generall

7、ywhenyou’reabouttoconsummatearetailpurchase-theyappearatyourshoulder,whisperingsuggestivelyinyourear.Amazonwasthepioneerofautomatedrecommendations,butNetflix,Apple,YouTubeandTiVohavethemtoo.Inthemusicspacealone,Pandorahasdozensofcompetitors.Agoodrecommendationengineisworthaloto

8、fmoney.AccordingtoareportbyindustryanalystForrester,on

当前文档最多预览五页,下载文档查看全文

此文档下载收益归作者所有

当前文档最多预览五页,下载文档查看全文
温馨提示:
1. 部分包含数学公式或PPT动画的文件,查看预览时可能会显示错乱或异常,文件下载后无此问题,请放心下载。
2. 本文档由用户上传,版权归属用户,天天文库负责整理代发布。如果您对本文档版权有争议请及时联系客服。
3. 下载前请仔细阅读文档内容,确认文档内容符合您的需求后进行下载,若出现内容与标题不符可向本站投诉处理。
4. 下载文档时可能由于网络波动等原因无法下载或下载错误,付费完成后未能成功下载的用户请联系客服处理。