Research Of Marketing Behavior And User Preferences Based On Weibo

With Weibo as a new media developed in China, Weibo marketing raised rapidly and attracted attention from all walks of life. Marketing based on Weibo can not only caused the majority of media people’s attention, but also attracted the favor of academia, for that the huge amount of data as a result of marketing activities on Weibo may be an appropriate scene to verify the advantage of data mining techniques. However, in the domestic researches on Weibo marketing are mainly concentrated in the media field currently. Based on this background, we analyze Weibo marketing behavior characteristics with the help of data mining algorithms in a more thorough and systematic view.This paper researches movie marketing behavior and user preferences based on Weibo. Firstly, we introduce a text-based topic extraction algorithm called TEBF, which based on FCM algorithm. In TEBF, we propose a new index called M-membership that is proved to be more suitable to indicate the importance of each term in this scenario, then use fuzzy set to represent clusters and topics, which can lead to a more reasonable and thematic result. Comparative experiments with K-Means and LDA illustrate our ability to cluster and extract topics in the field of Weibo. Then combined with the specific needs of Weibo marketing, the article proposes an analysis model of marketing behavior and user preference. From two aspects as Weibo content and user information, we give a comprehensive and systematic analysis based on the topic extraction algorithm and classification algorithm.In order to give ordinary users a more vividly view of the results, a Weibo marketing behavior analysis system called "WMA" has been developed, in terms of providing a convenient and effective platform for more users, which also proves the practical value of our study.