摘要
电子商务的飞速发展使网购成为社会消费的主要手段.借助八爪鱼数据爬虫软件,抓取京东商城运动鞋客户在线购买信息,统计分析消费者最常购买运动鞋的价格、颜色等;并基于商品评论构建在线评论贝叶斯网络结构模型,采用交叉验证方法检验模型,计算出相关节点的后验概率分布和条件概率分布,最终得出以下结论:客户满意度与其他相关变量间均存在明显的因果关系,且不同变量间也会存在关联;舒适度是提高客户满意度的关键因素.
The rapid development of e-commerce makes online shopping the main means of social consumption.This article uses the Octopus data crawler software to capture the online purchase information of Jingdong Mall sneakers customers,statistically analyze the prices and colors of the most commonly purchased sneakers by consumers;and build an online review Bayesian network structure model based on product reviews.The cross-validation method tests the model,calculates the posterior probability distribution and conditional probability distribution of the relevant nodes,and finally draws the following conclusions:there is a clear causal relationship between customer satisfaction and other related variables,and there will be correlations between different variables;Comfort is a key factor in improving customer satisfaction.
作者
沈长霞
车万留
桂海霞
SHEN Chang-xia;CHE Wan-liu;GUI Hai-xia(School of Econormics and Management,Anhui University of Science and Technology,Huainan 232001,China)
出处
《数学的实践与认识》
北大核心
2020年第23期285-294,共10页
Mathematics in Practice and Theory
基金
国家自然科学基金(61703005)
安徽理工大学教学研究项目(2017XJJY90,2018XJJY35)
安徽理工大学硕博基金项目(12059)。
关键词
数据挖掘
贝叶斯网络
在线评论
交叉验证
满意度
data mining
bayesian network
online reviews
cross-validation
satisfaction