社交网络平台用户特征识别及其行为演变--以微博为例

Users Feature Recognition and Behavior Evolution on Social Network Platforms--taking Weibo as an example

根据微博用户特征进行聚类,识别不同类别的用户特征;依据行为数据,分析用户不同使用行为的差异。对于优化社交网络平台、推进用户持续使用等具有重要意义。使用八爪鱼采集器抓取微博平台1200个用户的样本数据以及13840个微博数据,运用数理统计方法进行统计分析和归纳,探讨用户个人特征和行为规律。研究结果显示,微博用户无论在用户特征变量还是行为变量方面,均存在高度相关性。通过聚类结果进行不同类别用户特征的比较。在微博使用中发现持续使用、中较使用、停止使用等用户不同使用行为以及动态特征。

According to the characteristics of Weibo users,classify and identify different types of users'characteristics;According to the behavior data,analyze the differences of different user behaviors.It is of great significance to optimize the social network platform and promote the continuous use of users.The Octopus collector was used to capture the sample data of 1200 users and 13840 Weibo data of the Weibo platform,and the statistical analysis and induction were carried out using mathematical statistical methods to explore the personal characteristics and behavior rules of users.The research results show that Weibo users are highly correlated in both user characteristic variables and behavior variables.The characteristics of different categories of users are compared through clustering results.In the use of Weibo,we found different user behaviors and dynamic characteristics such as continuous use,dropout use,and stop use.