基于自适应免疫算法的网站广告分配优化  

Web Advertisement Allocation Optimization Based on Adaptive Immune Algorithms

作　　者：闫涛[1] 闫继涛[1] 

机构地区：[1]河南城建学院信息中心,河南平顶山467036

出　　处：《计算机工程与科学》2012年第1期183-186,共4页Computer Engineering & Science

摘　　要：如何优化网站广告分配,实现网站运行商和广告客户收益的双赢,是网站广告分配问题的关键。针对网站广告的特点,本文提出了一个基于混合定价策略的网站广告资源配置优化模型,将其建模为一个最大化网站总收益的约束优化问题,并通过自适应免疫克隆算法进行求解。自适应特性主要表现在:根据亲和度动态分配记忆单元和一般抗体单元,从而将全局搜索和局部搜索有效结合起来,有效提高了求解精度和收敛速度。同时,根据网站广告分配问题特点,设计了其它相关的免疫算子,如编码、克隆变异、克隆选择。仿真结果表明,算法有效平衡了网站广告客户和网站运营商的利益,求解结果更加实用。It is the key to web advertising that how to optimize its distribution to benefit both the web owner and advertisers. According to the properties of web advertisement, a hybrid pricing strategy based model is proposed. It is modeled as a constrained optimization problem to maximize the total revenue of the web. An adaptive immune algorithm is proposed to solve it. The adaptive characteristics lie in that, according to antibody affinity, antibody populations are dynamically divided into memory antibody units and general antibody units. It combines global search with local search effectively and hence improves the solution accuracy and the convergence rate. According to the characteristics of the solving problem, the relative immune operators are designed. The simulation results show that, the algorithm balances the revenues between the advertisement clients and web owners and the results are more practical.

关 键 词：网站广告 约束优化 自适应免疫克隆 混合定价 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程] TP393[自动化与计算机技术—控制科学与工程]