广告营销大数据的分析方法与应用分析研究 市场营销专业VIP免费

摘要随着科技的发展，广告行业也随之变化。从广告的媒介购买上来说，从传统的购买方式逐渐转向程序化购买这样更加科学精准的购买方式，但在现在这样的过渡时期，广告媒介代理公司仍旧面临着耗费大量的人力与物力在这样如果使用计算机就会迎刃而解的问题上。目前绝大部分广告方面的研究都专注于未来的程序化购买方面，而可以解决现在媒介公司所遇到问题的研究却少之又少。本研究建立了一个广告媒介的推荐系统。该系统通过专业数据公司提供的数据，利用时间序列预测技术，较为精准地预测手机端app的数据，再结合多维度的权威数据进行广告媒介的推荐。并经过实验证明，该系统有着较好的准确性和可信性。本研究为广告行业的去人工化、方便化、提供了一定的贡献，同时可以作为决策支持工具，对于媒介从业人员的工作（广告媒介选择）提供了相应的辅助。关键词：广告；时间序列预测算法；协同过滤算法；数据挖掘AnalysisandApplicationofBigDatainAdvertisementMarketingAbstractWiththedevelopmentofscienceandtechnology,theadvertisingindustryhaschanged.Intheadvertisingmediabuying,itchangesfromthetraditionalwaygraduallytotheprogrammaticpurchaseofsuchamorescientificandaccuratewayofpurchase,butinthistransitionalperiod,advertisingmediaagenciesstillfaceahugeamountofmanpowerandmaterialresourcesinthiswayiftheuseofcomputerswillbesolved.Atpresent,mostoftheresearchonadvertisingisfocusedonthefutureofprogrammaticpurchase,andalmostnoresearchcansolvetheproblemofmediacompanies.Thisstudyestablishedarecommendationsystemforadvertisingmedia.Basedonthedataprovidedbyprofessionaldatacompanies,thesystemusestimeseriespredictiontechnologytopredictthedataofmobilephone-endappmoreaccurately,andthencombinetheauthoritativedataofmulti-dimensionaltocarryontherecommendationofadvertisementmedium.Theexperimentalresultsshowthatthesystemhasgoodaccuracyandcredibility.Thisresearchprovidesacertaincontributionfortheadvertisingindustrytobeartificial,convenient,andcanbeusedasadecisionsupporttoolfortheworkofmediapractitioners(advertisingmediaselection)toprovideacorrespondingassistance.KeyWords：Advertisement；TimeSeriesPredictionMethod；CollaborativeFiltering；DataMining目录摘要.........................................................................................................................................IAbstract......................................................................................................................................II文献综述..................................................................................................................................11数据获取及预处理...............................................................................................................111.1APP表现数据.......................................................................................111.2移动端广告点位刊例............................................................................111.3移动端广告历史投放数据....................................................................121.4广告主数据............................................................................................121.5数据预处理............................................................................................132时间序列预测分析方法的研究...........................................................................................142.1时间序列数据挖掘....................................................................................................142.1.1数据挖掘基本概念.....................