摘要
从研究在线评论效用的影响因素入手,建立评论效用指标体系。采用模糊层次分析法确定指标的相对权重,通过语义挖掘对评论内容的各项指标进行量化处理,最后统计每条评论的效用总分。模型应用部分选取国内淘宝商城某商品的近2 000条商品评论信息进行实证分析。研究对比发现,经过排序模型处理后,大量的无用评论被后置,新排序中靠前的评论内容信息含量非常丰富,评论效用较高,能够有效地辅助其他消费者进行购物决策。
On the basis of studying the influencing factors of online reviews effectiveness, a review effectiveness index system is established. The fuzzy analytic hierarchy process is adopted to determine the relative weight of indexes, various indexes of reviews content are quantized by semantic mining, and the total effectiveness score is calculated for each review. In terms of the model application of this study, nearly 2 000 reviews on a product of China＇ s Tmall are selected to make an empirical analysis. The study and comparison indicates that, after being processed by the sequencing model, a large number of useless reviews are postponed, and those reviews at the forefront of the new sequence are very rich in information content and high in effectiveness, and can assist consumers in making purchase decisions effectively.
出处
《现代图书情报技术》
CSSCI
北大核心
2013年第4期62-68,共7页
New Technology of Library and Information Service
基金
国家大学生创新性实验计划(A类)基金项目"本地化电子商务平台的发展机制及其优化研究"(项目编号:A00750)的研究成果之一
关键词
信息挖掘
在线评论
效用排序
Information mining Online reviews Effectiveness sequencing
