摘要
设计一个迭代的MapReduce并行计算工作流,用于分析快消品电商网站的搜索引擎日志。该工作流根据每次检索在商品品牌字段上的层面搜索结果,挖掘关键字检索和品牌检索热度之间的潜在相关性,为关键字检索计算出其对品牌层面搜索结果集中各品牌的检索热度贡献值,最后对品牌检索热度贡献值列表进行归并计算得到各个品牌的检索热度排名并取Top-N。
An iterative MapReduce parallel computing workflow was designed to analyze the log of search engine in FMCG e-commerce websites.According to the facet search results on brand field of each retrieval,the workflow mines potential relevance between each keyword retrieval and the retrieval popularity of the brand,and calculates each keyword retrieval's contribution to the retrieval popularity of each brand in brand facet search results.Finally,the list of popularity contribution values of brand retrieval is combined and calculated to get the retrieval popularity ranking of each brand and the Top-N.
作者
王晨阳
WANG Chenyang(School of Information Science and Engineering,Fujian University of Technology,Fuzhou 350118,China)
出处
《福建工程学院学报》
CAS
2019年第4期365-370,共6页
Journal of Fujian University of Technology
基金
福建工程学院青年基金项目(GY-Z18168)
