关键词广告预算分配策略研究

发布时间：2018-03-02 03:24

本文关键词： 关键词广告预算分配策略模糊优化自适应变异粒子群算法差分进化算法　出处：《广西大学》2017年硕士论文　论文类型：学位论文

【摘要】：关键词广告是一种在线营销方式,因操作简单、预算可控以及广告效益较高等优点而得到了广泛的推广,迅速成为了广告商推广产品、投放广告的主要营销手段之一,同时也是近来国内外学者研究的热点之一.本文在已有研究的基础上,针对广告商投放广告时常选择的三个组合:单搜索引擎多关键词、多搜索引擎单关键词以及多搜索引擎多关键词,展开了系列的关键词广告预算分配策略的研究.(一)研究了单搜索引擎多关键词多阶段广告预算分配策略,在总预算限制下建立了以最大化广告收益、最小化无效点击为双目标的单引擎多关键词多阶段的广告预算分配模型,给出了模糊优化双目标预算分配模型的解法.实例验证了该模型的有效性与合理性.(二)研究了多搜索引擎单关键词多阶段广告预算分配策略,以单位收益所花费的竞价成本最小为单个搜索引擎的目标函数,构建了多搜索引擎多阶段的广告预算分配优化模型.针对模型易于陷入局部最优的缺陷,研究了用于求解该模型的自适应变异粒子群算法,探讨了在多个搜索引擎上竞价时的动态预算分配策略.实验表明,该方法具有更好的全局收敛性和实时性,有效地提高了广告商的广告收益.(三)研究了多搜索引擎多关键词多阶段广告预算分配策略,以每收益花费的成本和每预算产生的无效点击最小为目标,建立多搜索引擎多关键词多阶段的预算分配模型,采用了搜索能力较强的改进差分进化算法(IDE)对模型进行求解.实例验证,该算法具有较强的搜索能力以及较快的收敛速度.
[Abstract]:Keyword advertising is an online marketing method, which has been widely promoted because of its advantages of simple operation, controllable budget and high advertising efficiency. It has quickly become one of the main marketing means for advertisers to promote products and place advertisements. At the same time, it is also one of the hot spots of scholars at home and abroad recently. Based on the existing research, this paper focuses on the three combinations that advertisers often choose: single search engine and multiple keywords. Multi-search engine single keyword and multi-search engine multi-keyword, launched a series of keyword advertising budget allocation strategy. (1) single search engine multi-keyword multi-stage advertising budget allocation strategy, An ad budget allocation model with single engine, multi-keyword and multi-stage, which maximizes advertising revenue and minimizes invalid clicks, is established under the limit of total budget. In this paper, the solution of fuzzy optimization model for double objective budget allocation is given. The effectiveness and rationality of the model are verified by an example. (2) the multi-search engine single keyword and multi-stage advertising budget allocation strategy is studied. Taking the minimum bidding cost per unit income as the objective function of a single search engine, a multi-search engine multi-stage advertising budget allocation optimization model is constructed. The adaptive mutation particle swarm optimization algorithm for solving the model is studied, and the dynamic budget allocation strategy for bidding on multiple search engines is discussed. The experimental results show that the proposed method has better global convergence and real-time performance. Effectively improve the advertising revenue of advertisers. (3) study the multi-search engine multi-keyword multi-stage advertising budget allocation strategy, aiming at the cost per revenue and the minimum of invalid clicks per budget. A multi-search engine multi-keyword multi-stage budget allocation model is established and the improved differential evolution algorithm (IDE) is used to solve the model. The example shows that the algorithm has strong searching ability and fast convergence speed.
【学位授予单位】：广西大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.3;TP18

【参考文献】