论文部分内容阅读
关键字是搜索竞价广告中联结广告主、搜索用户和搜索引擎之间的重要纽带。关键字的生成、选择与分组等过程一直都是搜索竞价广告主所面临的核心决策问题,并贯穿于搜索竞价推广活动的整个生命周期。本文提出了一个针对搜索竞价推广活动整个生命周期的多层次、闭环关键字策略优化框架模型,分别从关键字的生成、扩展、定位、选择与组合以及动态调整几个方面研究相应的最优化策略。进而,利用实际关键字竞价推广活动的日志数据,设计了相应的搜索竞价关键字策略实验场景,对本文提出的关键字优化模型及其算法进行实际验证与评估。实验结果表明,本文所提出的关键字策略优化框架模型及其算法能够收敛且在很大程度上优于一般广告主所采取的关键字策略。
The keyword is an important link between search advertisers, search advertisers, and search engines in search auction ads. The process of generating, selecting, and grouping keywords has always been a core decision-making issue for search auction advertisers and runs through the search for the entire lifecycle of auction campaigns. This paper presents a multi-level, closed-loop keyword strategy optimization framework model for searching the entire life cycle of bidding promotion activities, and studies the corresponding optimization strategies from the aspects of keyword generation, expansion, positioning, selection and combination and dynamic adjustment . Furthermore, using the actual keyword bidding log data of promotional activities, we design a corresponding experiment scenario of bidding keyword strategy, and verify and evaluate the proposed keyword optimization model and its algorithm. The experimental results show that the proposed keyword strategy optimization framework model and its algorithm can converge and largely outperform the keyword strategies adopted by advertisers in general.