基于社会网络分析下文本挖掘的微博营销

发布时间：2018-04-22 10:13

本文选题：微博营销 + 社会网络分析　；参考：《兰州财经大学》2015年硕士论文

【摘要】：随着互联网络的发展,传统的营销渠道已经不能满足企业的需求。新媒体营销方式不断涌现,其中微博营销最为突出。主要由于微博平台的互动性和实时性,为企业提供了一个高效直接的营销渠道。本文以如何找到微博内容传播的关键用户以及该用户所发微博的主题为切入点,以真实的微博数据为支撑,运用社会网络分析方法找到微博传播的关键用户,同时运用文本挖掘的LDA模型(Latent Dirichlet Allocation,简称LDA),得到关键用户所发微博的主题内容。基于这两点分析的内容,为企业如何成为网络领袖以及微博营销的主题选取提供理论支持和实际建议。本文首先介绍了研究背景和选题意义,通过国内外学者对微博营销、基于社会网络分析的微博研究、文本挖掘的LDA模型3个方面的研究现状发现,目前并没有将两种方法结合使用的实际研究。然后,针对社会网络的含义、特征、主要要素,以及社会网络分析的相关概念和应用,做了较详细的研究;并且深入探讨了文本预处理技术中最重要的分词技术,和文本挖掘的LDA模型的相关理论。最后,通过因特网抓取微博数据,运用社会网络分析法,确定5位转发网络以及评论网络中共同存在的网络领袖;并且运用文本挖掘的方法,挖掘网络领袖所发微博以及评论内容的主题,最后根据这些主题为微博营销提出建议。基于社会网络分析找到关键用户并且挖掘该用户所发微博内容的主题,不仅可以有针对性的挖掘文本内容,而且可以降低文本挖掘LDA模型的运算成本。使挖掘营销主题的过程更为精确、高效。
[Abstract]:With the development of Internet, traditional marketing channels can not meet the needs of enterprises. New media marketing methods have been emerging, among which Weibo marketing is the most prominent. Mainly due to the interaction and real-time of Weibo platform, it provides an efficient and direct marketing channel for enterprises. Based on how to find the key users of Weibo's content dissemination and the theme of the user, and the actual data of Weibo, this paper uses the social network analysis method to find the key users. At the same time, the LDA model of text mining is used to obtain the theme content of Weibo sent by key users. Based on these two analyses, it provides theoretical support and practical advice on how to become a network leader and Weibo's marketing theme selection. This paper first introduces the research background and the significance of the topic, through the domestic and foreign scholars to Weibo marketing, social network analysis based on Weibo research, text mining LDA model of three aspects of the status quo. There is no practical study of the combination of the two methods. Then, the meaning, characteristics, main elements of social network, as well as the related concepts and applications of social network analysis, are studied in detail, and the most important word segmentation technology in text preprocessing technology is discussed in depth. And the theory of LDA model of text mining. Finally, through the Internet to grab Weibo data, using the social network analysis method, to determine the 5-bit forwarding network and comment network network co-exist network leaders, and the use of text mining method, Excavate the themes of Weibo and comments by internet leaders, and finally make suggestions for Weibo marketing based on these themes. Finding the key user and mining the subject of Weibo content based on social network analysis can not only mine text content, but also reduce the computing cost of text mining LDA model. Make the process of mining marketing themes more accurate and efficient.
【学位授予单位】：兰州财经大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：F274;G206-F

【参考文献】