来自网上用户评论的声音-基于文本数据的产品评价分析.docVIP

来自网上用户评论的声音 ——基于文本数据的产品评价分析 摘 要 互联网的迅猛发展使得用户生成内容数量呈爆炸性增长，国内利用用户生成内容，分析产品评价的文献如凤毛麟角；国外的相关研究局限于语言差异，无法在中国市场直接推广。 本文的贡献在于：首先，基于网上顾客产品评价的文本数据进行产品评价分析，而不是以问卷、深度访谈或焦点小组讨论,较好的削弱了顾客由于记忆偏差产生的调研误差；其次，从顾客对产品的关注点出发，而不是企业的关注点，在产品评价上更具顾客导向；再次，基于计算机切词技术和模糊综合评价法，本方法可对产品进行持续性评价，从而了解产品评价的动态变化。 关键词：用户生成内容；产品评价；文本数据；模糊综合评价法 ABSTRACT The number of user generated content has an explosive growth, due to the rapid development of Internet. This kind of phenomenon provides rich materials for the academic research, also a better way for enterprises to understand consumers’ demand and their feelings of product performance. However, there are only a few paper about how to get product evaluation from the user generated content in the domestic. Many foreign literatures about UGC cannot be used directly to evaluate products in china for the language differences. In the study, this paper builds a product evaluation system which is suitable for online comments. Then, to verify the feasibility of this method, we collected more than 500 online comments, and make an analysis of these data by using word segmentation technology and Fuzzy Comprehensive Evaluation. Finally, we compare our results with the product evaluation index on Taobao website, finding that our evaluation method is more detailed and more targeted. We are different with previous product evaluation method. First of all, this article is based on online customers product comments, rather than the questionnaire, in-depth interviews or focus group discussion; Secondly, we embarks from the customers product concerns, rather than the concerns of firms ; thirdly, based on computer word cutting technology and fuzzy comprehensive evaluation , our method can focus on consumers feeling of product performance over time ,so as to understand the dynamic changes of the product evaluation. Key words: User generated content; Product evaluation; text data; Fuzzy comprehensive evaluation 引言