基于在线品牌社区意见领袖的用户关键需求挖掘

Users’Key Demands Mining Based on Opinion Leaders in Online Band Community

随着社会的快速发展以及技术的不断进步,人们生活节奏不断加快,对产品的需求也在快速发生着变化。在线评论是目前用户需求表达的重要渠道。为克服不加区分挖掘所有评论的传统用户需求挖掘方法耗时过长,难以聚焦用户关键需求的问题,从用户关键性与需求关键性的双关键性角度出发,研发了一种基于在线品牌社区意见领袖的用户关键需求挖掘方法以快速获取用户重要需求,简称KEY-DEMANDS-OL。该方法依据帕累托法则,依托意见领袖评论大数据,采用优化的情感程度及初始改进率,结合KANO模型对用户关键需求进行挖掘。该方法不仅考虑了程度副词的语义信息,提高了情感分析的准确率,而且能够完成意见领袖的生成内容与KANO模型的自动整合,实现用户关键需求的获取与分类。研究结果表明,与挖掘所有评论的传统方法相比,KEY-DEMANDS-OL可以快速获取用户的关键需求,为企业制定产品优化方案提供辅助决策支持。

With the rapid development of society and the continuous progress of technology,the pace of people's lives is constantly accelerating.At the same time,people’s demand for products is also rapidly changing.Currently,the online reviews are the important channels through which the users can express their demands.To overcome the problem that the traditional user demand mining method of mining all comments without discriminating takes too long and is difficult to focus on the key needs of users,starting from both the key users and the key demands,a method for users’key demands mining based on the opinion leaders in online brand community is developed to quickly obtain the users’important demands,which can be called KEY-DEMANDS-OL for short.It discovers the users’key demands integrated with KANO model based on the reviews big data of opinion leaders according to the Pareto principle,which also improves the initial improvement rates and the emotional degree.This novel method increased the accuracy of emotional analysis by taking the semantics of degree adverbs into account.In addition,it can integrate the UGC of opinion leaders with the KANO model automatically and obtain the users’key demands and the category of the demands.The research showed that comparing to the traditional method that mining all the uses’reviews,the KEY-DEMANDS-OL can obtain the key demands of users quickly and provide auxiliary decision-making support for enterprises to develop product optimization plans.