产品质量和售后服务问题已成为决定企业发展和用户消费的关键因素。目前,产品舆情分析的主要研究方法都是依靠关键词从社交网络渠道获取消费者对产品的评价,这种方法由于缺乏对舆情语义的分析,会产生比较多的噪音数据。本文提出了一种基于产品舆情主题结构和产品舆情文法(称为PPOG)的产品舆情分析方法,研制了一种具有较高精度的产品舆情分析算法。通过大量实际的微博例子测试,基于PPOG的分析方法具有良好的应用前景。
Product quality and safety have been viewed as crucial factors to both enterprise development and customer concern. At present, keyword-based methods are the primary analysis method of product- relevant public opinion. However, due to the lack of a deep analysis of public opinions, numerous wrong results are produced. In this paper, a thematic structure of product public opinions is proposed and a product-public opinion grammar, called PPOG, is designed to analyze mircro-blogs. Through comprehensive experiments with the use of miero-blogs, it is believed that the PPOG-based analysis method of public opinion has a promising application prospect.