中文产品评论的“特征观点对”识别:基于领域本体的建模方法

发布时间：2018-01-29 17:59

本文关键词： 中文产品评论特征观点对领域本体评论挖掘模型　出处：《系统工程》2013年01期 　论文类型：期刊论文

【摘要】：随着社交媒体的发展,不断增加的在线产品评论正在极大地影响电子商务市场,使得评论挖掘成为商业界与学术界共同的热点话题。针对中文产品评论的特点,本文提出一种基于领域本体的建模方法,通过建立评论挖掘模型来对产品评论的基本评价单元——"特征观点对"进行识别。该建模过程以设计科学研究方法论为指导。首先在模型设计阶段,构建面向产品评论的领域本体;然后在模型实施阶段,提出基于本体的特征观点对识别方法;最后在模型评价阶段,通过实验对评论挖掘结果进行评价。实验结果表明,本文提出的方法与其他基于统计的方法以及基于语义的方法相比,在性能上有明显提高,对克服口语化严重和语法不规范等问题具有良好的效果。此外,通过特征观点对的识别与统计,使产品评论这种非结构化文本转化为机器可读的、能理解的结构化表达并得到具有一定商业应用价值的信息。
[Abstract]:With the development of social media, the increasing number of online product reviews is greatly affecting the e-commerce market, making comment mining become a common hot topic in business and academia. This paper presents a modeling method based on domain ontology. In this paper, a comment mining model is established to identify the basic evaluation unit of product review, "feature view pair". The modeling process is guided by the methodology of design scientific research. First, it is in the stage of model design. Constructing domain ontology for product review; Then, in the stage of model implementation, an ontology-based feature view pair recognition method is proposed. Finally, in the stage of model evaluation, the results of comment mining are evaluated through experiments. The experimental results show that the proposed method is compared with other statistical methods and semantic based methods. It has a good effect on overcoming the problems of serious colloquial and nonstandard grammar. In addition, the recognition and statistics of feature point of view are adopted. The unstructured text of product review is transformed into machine-readable and comprehensible structured expression and the information with certain commercial application value is obtained.
【作者单位】：同济大学经济与管理学院;井冈山大学商学院;
【基金】：国家自然科学基金资助项目(70971099) 中央高校基本科研业务费专项(1200219198) 上海市科技发展基金软科学研究博士生学位论文资助项目(12692193000)
【分类号】：TP391.1;F49;F224
【正文快照】： 1引言在线评论反映了用户通过互联网对产品各方面发表的看法。与商家的促销信息相比,在线评论具有独立性、非商业性,因此深得用户信赖。与此同时,由于缺少线下体验,更多的用户倾向于先看评论,后做决策。DeloitteConsumer Products Group调查显示,有67%的网民会浏览在线评论,

【相似文献】