首页> 外文学位 >Digital advertising ecosystem: Prescriptive analytics for novel operational and strategic challenges.
【24h】

Digital advertising ecosystem: Prescriptive analytics for novel operational and strategic challenges.

机译:数字广告生态系统:针对新型运营和战略挑战的规范分析。

获取原文
获取原文并翻译 | 示例

摘要

The digital advertising phenomenon has resulted in an entire ecosystem of supply-side and demand-side players, as well as intermediaries that serve to mitigate supply-demand inefficiencies. My dissertation provides prescriptive solutions to three novel operational and strategic challenges faced by different stakeholders in this ecosystem.;The first problem arises from the misalignment of incentives between a supply-side network (a firm that monetizes events of users' visits for a diverse network of websites) and an ad aggregator (a firm that buys events of users' visits to websites or mobile applications on behalf of advertisers). This results in a situation where the supply-side network is left to optimize the click-revenue generated from the traffic (i.e., impressions) it sends to the ad aggregator without knowledge of the objective function. I develop an effective data-driven approach to solve constrained optimization problems in which the decision maker does not have an analytical form for the objective function, but knows what decision variables affect the function.;The second problem arises from the interaction between a supply-side network and advertisers. The challenge faced by the supply-side network is to manage fading ads (or fads) shown to an end user during a session of a mobile application (app). In addition to the question of what ads to show, this ad-sequencing problem also seeks the sequence in which the ads are displayed and the durations for which they are shown. I formulate this problem as a stochastic dynamic program in which uncertainty arises from two sources: the duration for which the user stays on the app and her clicking behavior. Optimal or near-optimal policies for a variety of practical settings are developed.;The third problem originates from the interaction between an ad exchange (a supply-demand matching platform where impressions are sold through real-time auctions) and its participating ad aggregators (bidders). For an impression that is being auctioned by the exchange, the ad aggregators respond with bids based on the information they receive (or possess) about that impression. I examine the impact of impression-related information: How much information should an ad exchange reveal? and a related question: Can ad exchanges make more money by hiding some pertinent attributes of impressions (e.g., identity of the publisher)? I develop good (and simple) information revelation designs and policies for ad exchanges. The welfare implications of these policies on the other stakeholders are also analyzed.
机译:数字广告现象已经形成了供方和需求方参与者的整个生态系统,以及有助于减轻供需效率低下的中介机构。我的论文为该生态系统中不同利益相关者面临的三个新的运营和战略挑战提供了规范性的解决方案。第一个问题是由于供应方网络(将用户访问事件货币化为多样化网络的公司)之间的激励机制不匹配而引起的。的网站)和广告聚合器(一家代表广告商购买用户访问网站或移动应用程序的事件的公司)。这导致一种情况,即供应方网络留给最优化从它发送给广告聚合器的流量(即印象数)生成的点击收入,而无需了解目标函数。我开发了一种有效的数据驱动方法来解决约束优化问题,其中决策者没有目标函数的解析形式,但是知道哪些决策变量会影响函数。第二个问题是由供应商之间的相互作用引起的。侧网络和广告客户。供应方网络面临的挑战是管理在移动应用程序(app)会话期间显示给最终用户的淡入淡出广告(或淡入淡出)。除了显示什么广告的问题之外,此广告排序问题还寻求显示广告的顺序和显示时间。我将此问题描述为一个随机动态程序,其中不确定性来自两个来源:用户停留在应用程序上的持续时间和点击行为。制定了针对各种实际情况的最优或接近最优的政策;第三个问题来自广告交易平台(通过实时拍卖出售印象的供需匹配平台)与其参与的广告聚合商之间的相互作用(投标人)。对于由交易所拍卖的印象,广告聚合者基于他们收到(或拥有)的关于该印象的信息来出价。我研究了与展示相关的信息的影响:广告交易应显示多少信息?还有一个相关的问题:通过隐藏一些相关的展示属性(例如,发布商的身份),广告交易平台能否赚到更多的钱?我为广告交易开发了良好(简单)的信息展示设计和政策。还分析了这些政策对其他利益相关者的福利含义。

著录项

  • 作者

    Sun, Zhen.;

  • 作者单位

    The University of Texas at Dallas.;

  • 授予单位 The University of Texas at Dallas.;
  • 学科 Management.;Marketing.
  • 学位 Ph.D.
  • 年度 2016
  • 页码 190 p.
  • 总页数 190
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 康复医学;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号