基于产品评论的消费者情感波动分析模型构建及实证研究
发布时间：2018-07-23 20:52
【摘要】：【目的】基于网络产品评论信息构建消费者的情感波动分析模型。【方法】该分析模型以文本挖掘技术理论为基础,运用产品评论挖掘技术和情感分析技术,在情感分析的同时充分考虑不同类型连词对句子情感倾向性的影响,并采用相应的权值计算方法。【结果】从京东和中关村在线抓取某款手机从2013年11月到2015年1月这段时间内产品评论信息并进行分析,验证了该模型的有效性。【局限】在分析消费者情感波动主要影响因素方面,该分析模型主要考虑相邻时间段内产品特征词个数的变化以及产品特征词在评论信息中出现次数的变化这两个维度,其他维度并未涉及。【结论】该模型有效地分析了消费者在一段时间内的情感波动趋势,以及产生情感波动的因素,能够为企业决策提供一定的参考。
[Abstract]:[objective] to construct a consumer emotion fluctuation analysis model based on product review information. [methods] based on the theory of text mining, the model uses product review mining technology and emotion analysis technology. At the same time of affective analysis, the influence of different types of conjunctions on the affective tendency of sentences is fully considered. Using the corresponding weight calculation method. [results] grab the product comment information from JingDong and Zhongguancun online from November 2013 to January 2015 and analyze it. The validity of the model is verified. [limitations] in terms of analyzing the main influencing factors of consumer emotional fluctuation, The analysis model mainly considers the two dimensions of the number of product feature words in adjacent time periods and the number of product feature words appearing in the comment information. Other dimensions are not involved. [conclusion] this model can effectively analyze the trend of consumer emotion fluctuation in a certain period of time, and the factors that produce emotional fluctuation, which can provide a certain reference for enterprises to make decisions.
【作者单位】： 宁波大学机械工程与力学学院;
【基金】：浙江省自然科学基金项目“中小企业业务问题及其解决方案的建模理论与管理方法研究”(项目编号:LY14G010003) 国家星火计划项目“农业业务问题的大数据推荐技术应用与示范”(项目编号:2015GA701011) 国家自然科学基金项目“面向产业集群广义制造系统的知识模块化机理研究”(项目编号:71671097)的研究成果之一
【分类号】：TP391.1
，
本文编号：2140607
[Abstract]:[objective] to construct a consumer emotion fluctuation analysis model based on product review information. [methods] based on the theory of text mining, the model uses product review mining technology and emotion analysis technology. At the same time of affective analysis, the influence of different types of conjunctions on the affective tendency of sentences is fully considered. Using the corresponding weight calculation method. [results] grab the product comment information from JingDong and Zhongguancun online from November 2013 to January 2015 and analyze it. The validity of the model is verified. [limitations] in terms of analyzing the main influencing factors of consumer emotional fluctuation, The analysis model mainly considers the two dimensions of the number of product feature words in adjacent time periods and the number of product feature words appearing in the comment information. Other dimensions are not involved. [conclusion] this model can effectively analyze the trend of consumer emotion fluctuation in a certain period of time, and the factors that produce emotional fluctuation, which can provide a certain reference for enterprises to make decisions.
【作者单位】： 宁波大学机械工程与力学学院;
【基金】：浙江省自然科学基金项目“中小企业业务问题及其解决方案的建模理论与管理方法研究”(项目编号:LY14G010003) 国家星火计划项目“农业业务问题的大数据推荐技术应用与示范”(项目编号:2015GA701011) 国家自然科学基金项目“面向产业集群广义制造系统的知识模块化机理研究”(项目编号:71671097)的研究成果之一
【分类号】：TP391.1
，
本文编号：2140607
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2140607.html
