在线口碑传播的意见领袖识别方法研究

Web2.0技术应用的不断深化,使得越来越多的用户在网络上发布对产品、服务、品牌或企业的评价,这些在线口碑极大影响着消费者的购买决策和企业的声誉、利润,如何识别出在线口碑传播中的意见领袖对企业有着重要价值。本文结合RFM模型和情感词自动判别方法,考虑情感(S)这一个新的指标,提出了RFMS模型来测量在线口碑发布者的影响力,应用人工神经网络识别出意见领袖,以程度中心性来评估识别结果的有效性。本文采用网络上获取的大众点评网的美食评论语料,对所提出的方法进行了数据实验,实验结果表明,该方法能够精确识别出意见领袖。

Study on the Method of Identifying Opinion Leaders for Online Word-of-mouth Communication

The deepening adoption of Web2.0 technology makes more and more users publish reviews on the web about products, services, brands or business, and the online word-of-mouth greatly influences customers’ purchasing decisions and corporate reputations. Therefore, it makes sence for enterprises to identify the opinion leaders of online word-of-mouth communication. In this study, combining RFM model and automatically measuring method of sentiment words, a RFMS model is proposed to measure the influence of online word-of-mouth publisher, identify opinion leaders by applying artificial neural network, and assess the validity of identifying results based on degree centrality. The online reviews of dianping.com are analyzed, and based on those reviews the proposed method is verified. Results show that the proposed method in this paper can accurately identify opinion leaders.