基于数据挖掘的电商搜索广告投放策略研究

An E-commerce Searching Advertisement Serving Strategy Based on Data Mining

为了降低商家的广告投入成本,同时优化用户在电商平台中的购物体验,本文应用阿里"天池大数据"平台提供的海量真实数据,构建并训练出以广告转化率为预测目标的机器学习模型。通过分析模型学习结果,挖掘出搜索广告转化率的影响因素及影响效应。结果表明:店铺物流服务、广告商品销量、消费者偏好、电商平台智能化程度等对转化率的影响较大。同时对所有影响因素在用户、店铺、广告商品、广告上下文4个维度进行归类和分析。据此提出基于平台视角及商家视角下的广告投放策略,为电商平台及其入驻商家提供参考。

In order to reduce the cost of advertising investment for businesses, and optimize the shopping experience of users in the e-commerce platform, the massive real data provided by Ali “Tianchi Big Data” platform is applied and a machine learning model is built and trained with the target of advertising conversion rate. By analyzing the model learning results, the influencing factors and effects of search advertising conversion rate are explored. The results show that store logistics services, advertising product sales, consumer preferences, and the degree of intelligence of e-commerce platforms have a greater impact on conversion rates. At the same time, all the influencing factors are classified and analyzed in the four dimensions of users, shops, advertising products and advertising contexts. Based on this, the advertisement serving strategy based on the platform and the merchant perspectives is proposed to provide reference for the e-commerce platform and its resident merchants.