Skip to main content
QUICK REVIEW

[论文解读] Display Advertising with Real-Time Bidding (RTB) and Behavioural Targeting

Jun Wang, Weinan Zhang|arXiv (Cornell University)|Oct 7, 2016
Consumer Market Behavior and Pricing被引用 14
一句话总结

本专著对展示广告中的实时出价(RTB)提供了全面的技术综述,将行为定向与机器学习相结合,用于用户响应预测、出价行情预测和动态定价。它提出了可扩展的实时出价算法、欺诈检测方法以及归因建模技术,实现了自动化、数据驱动的广告优化,并在点击率和转化率方面实现了可衡量的性能提升。

ABSTRACT

The most significant progress in recent years in online display advertising is what is known as the Real-Time Bidding (RTB) mechanism to buy and sell ads. RTB essentially facilitates buying an individual ad impression in real time while it is still being generated from a user's visit. RTB not only scales up the buying process by aggregating a large amount of available inventories across publishers but, most importantly, enables direct targeting of individual users. As such, RTB has fundamentally changed the landscape of digital marketing. Scientifically, the demand for automation, integration and optimisation in RTB also brings new research opportunities in information retrieval, data mining, machine learning and other related fields. In this monograph, an overview is given of the fundamental infrastructure, algorithms, and technical solutions of this new frontier of computational advertising. The covered topics include user response prediction, bid landscape forecasting, bidding algorithms, revenue optimisation, statistical arbitrage, dynamic pricing, and ad fraud detection.

研究动机与目标

  • 解决实时展示广告中自动化、可扩展且数据驱动优化日益增长的需求。
  • 通过将RTB视为一个动态的多智能体学习问题,弥合信息检索、数据挖掘与机器学习之间的信息鸿沟。
  • 通过行为数据与预测建模,实现精准的用户级定位与出价优化。
  • 通过整合统计套利、动态定价与欺诈检测机制,提升广告投放效果。
  • 为计算广告与机器学习领域的研究人员和从业者提供统一的技术框架。

提出的方法

  • 采用逻辑回归、因子分解机及集成方法(如梯度提升树)进行实时用户响应预测。
  • 应用基于树的对数正态模型、右删失线性回归与生存模型,预测出价行情与中签概率。
  • 采用诚实出价与线性出价策略,并结合预算约束的优化方法,分别针对点击与转化进行优化。
  • 通过多智能体学习与后悔最小化机制,在多个广告活动之间挖掘统计套利机会。
  • 应用Shapley值与概率模型,实现基于数据驱动的转化路径归因。
  • 采用共访网络分析与可视性指标,检测广告欺诈并提升活动完整性。
Figure 2.1: The various players of online display advertising and the ecosystem: 1. The advertiser creates campaigns in the market. 2. The market trades campaigns and impressions to balance the demand and supply for better efficiency. 3. The publisher registers impressions with the market. 4. The us
Figure 2.1: The various players of online display advertising and the ecosystem: 1. The advertiser creates campaigns in the market. 2. The market trades campaigns and impressions to balance the demand and supply for better efficiency. 3. The publisher registers impressions with the market. 4. The us

实验结果

研究问题

  • RQ1如何利用流式出价请求数据,在实时环境中准确预测用户响应概率?
  • RQ2在RTB拍卖中,何种出价策略可在遵守预算约束的前提下,最大化转化或点击?
  • RQ3出价行情预测如何提升出价决策质量,并减少在第二价格拍卖中的过度支付?
  • RQ4迁移学习在基于网络浏览行为提升点击预测性能方面发挥何种作用?
  • RQ5基于Shapley值或概率框架的归因模型如何改善广告活动绩效的度量?

主要发现

  • 集成模型(如梯度提升树)及混合模型在用户点击预测方面优于传统逻辑回归。
  • 诚实出价是第二价格拍卖中的主导策略,验证了在RTB中采用诚实出价机制的有效性。
  • 生存模型与右删失线性回归在出价行情预测中表现稳健,相比基线方法具有更高的准确性。
  • 跨广告活动的统计套利挖掘可识别出具有可衡量投资回报率提升的盈利出价机会。
  • 共访网络分析能有效检测由僵尸网络驱动的欺诈行为,通过识别异常的用户访问模式。
  • 基于自助抽样逻辑回归的数据驱动归因模型可改善转化路径分析,并支持更优的预算分配。
Figure 2.2: How RTB works for behavioural targeting. Source: [ Zhang et al., 2014b ] .
Figure 2.2: How RTB works for behavioural targeting. Source: [ Zhang et al., 2014b ] .

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。