基于相关性模型与大数据分析的关键词优化方法研究

Research on Keyword Optimization Method Based on Correlation Model and Big Data Analysis

随着计算机网络与信息技术的快速发展,电子商务得到了爆发式的发展,电子商务网店运营成为广大电商从业者所追捧的技术。随着大数据技术的普及应用,电子商务网店运营技术也在不断地推陈出新,基于大数据的关键词筛选优化就是人们所关注的重要技术。相关性模型是电商平台商品搜索排序的基础模型,基于相关性模型的搜索引擎优化(SEO)是目前电子商务数据化运营中进行关键词优化筛选的基本方法。该文将原有方法与大数据分析相结合,从关键词采集、筛选、组合、监测、调换等多维度进行关键词的优化筛选,提出并论述关键词设置的三个筛选优化模型:竞争系数、单品位竞争个数、关键词主体价格区间适应性,并对筛选优化后选用的关键词进行了应用有效性实践监测。大量监测数据表明,使用相关性模型与大数据分析相结合的关键词筛选优化方法得到的商品标题关键词,可以有效提升商品的曝光量、点击率与转化率,实现更高效的网店引流与成交量。

With the rapid development of computer networks and information technology,e-commerce has experienced explosive development.E-commerce online store operation has become a technology pursued by a large number of e-commerce practitioners.With the popularization and application of big data technology,e-commerce online store operation technology is also constantly evolving,and keyword selection and optimization based on big data is an important technology that people pay attention to.The correlation model is the basic model for sorting product searches on e-commerce platforms.Search Engine Optimization(SEO)based on the correlation model is currently the basic method for keyword optimization and screening in e-commerce data-driven operations.We combine the original method with big data analysis to optimize and screen keywords from multiple dimensions such as keyword collection,screening,combination,monitoring,and exchange,and propose and discuss three optimization models for keyword selection:competition coefficient,number of single grade competitions,and adaptability of keyword subject price range.Conduct practical monitoring on the effectiveness of the selected keywords after selection optimization.A large amount of monitoring data shows that using the keyword selection optimization method combining correlation model and big data analysis to obtain product title keywords can effectively improve the exposure of the product click through rate and conversion rate,achieving more efficient online store traffic and transaction volume.