摘要
[目的 /意义]对在线虚假评论的现有研究进行梳理,分析研究现状,明确未来研究发展方向。[方法 /过程]以CNKI和Web of Science文献为研究对象,从文献分析的视角,采用定性与定量分析相结合的方法,从虚假评论的识别方法、特征提取以及防治策略的角度,对国内外虚假评论研究的现状进行了分析,总结和概括了本领域研究的热点和存在的问题。[结果/结论]研究表明,在虚假评论的识别方法上,需加强对半监督和无监督学习的研究;在特征提取上,可考虑本体技术的应用;在防治策略上,要考虑多学科多领域的合作。
[ Purpose/significance] The paper carded existing study about online product opinion spam, summarized re- search status, and put forward future research direction. [ Method/process] From the perspective of document analysis, using a combination of qualitative and quantitative methods to analyze the present status of research on opinion spares, summarized the key issues and problems in the field. [ Result/conclusion] This paper presented the existing problems and future direction of field from three aspects： recognition method, feature extraction and prevention strategy, which provided reference to the academic research more deeply.
作者
朱娟
Zhu Juan(School of Information Management, Wuhan University, Wuhan 430072, China School of Information, Jiujiang University, Jiujiang 332005, China)
出处
《现代情报》
CSSCI
北大核心
2017年第5期166-171,共6页
Journal of Modern Information
基金
国家自然科学基金资助项目项目"社会化媒体集成检索与语义分析方法研究"(项目编号:71273194)
关键词
在线商品
虚假评论
文献分析
机器学习
识别方法
防治策略
特征提取
综述
online products
opinion spare
literature analysis
machine learning
recognition method
prevention strategy
feature extraction
