资源描述：
《python和mysql实现Yelp数据分析》由会员上传分享，免费在线阅读，更多相关内容在应用文档-天天文库。
1、RunningHead:YELPDATASETANALYSISYelpdatasetanalysisXiaoyuChen(20663878)WangYang(20663316)UniversityofWaterlooWinter2017ECE656DatabaseSystemYELPDATASETANALYSIS2AbstractTheaimoftheprojectistoanalyzetheYelpdataset.TherearealargequantityintheYelpdataset.Inor
2、dertoanalyzetheYelpdatasetbasedonvaliddata,wecompletedtodatacleaningtoremovingormodifyingtherecordswhicharenotconformtosomeconstraints.Besides,weaddedtheindexafterimportingtheserecordsintodatabasetofindthedifferencewithindexandwithoutindex.Weselectedtwo
3、topicstoanalyze.Oneofthemistopredicttheratinggivenbyausertoabusinesswheretheuserhasneverbeentherebefore.Weanalyzedthedistributionofuser’sratingandthereasoninwhichtheygavetheseratings.Thesecondanalysisistofindifabusinessisdecliningorimprovinginitsratinga
4、ndtriedtofindthetrendofthechangeofratings.Keyword:datacleaning,dataindexing,yelp,databaseYELPDATASETANALYSIS3YelpdatasetanalysisIntroductionYelpdatasetisanonlinedatabaseprovidedbyofficialwebsitewhichcontainstheinformationofbusinessesandtheirusers.Thedat
5、asethasfivedifferentdatatypes:businesses,reviews,users,checkinsandtips.Thebusinessesstoretheinformationofeverybusiness.Forexample,eachbusinesshasanuniquebusinessid.Anditalsohasname,fulladdress,locationinformationincludinglatitudesandlongitudes,hoursando
6、therkindsofinformation.Thereisareview_countwhichrepresentsthenumberofreviewsthatthebusinessgot,andanaveragestarratingcalculatedaccordingtoallthereviews.Thedatatypereviewsconsistofbusiness_id,user_id,starratingwhichcanbejustifiedbythecorrespondingreviewt
7、ext.Andeveryreviewcangetvotesfromotherusers,iftheyfindthereviewisuseful,funnyorcool.Yelpisanapplicationtohelpusersfindplacestheywouldliketovisit.Ithasalargeamountinformationrelatedtoavastvarietyofbusiness,likerestaurants,bras,cafes,hotelsandsoon.However
8、,italsofocusonusers.Thedatatypeusersnotonlystorethebasicinformationaboutthemliketheirnames,howlongtheyhavebeenusingYelpfor,theyearofbeingeliteandthenumberofreviewstheymade,butalsoprovidetheopportunityforuserstobecomesfriendsorfan
