基于社交网络数据的用户群体画像构建方法研究

Building user group portraits based on social networking data

随着网络技术的发展,社交网络成为人们信息获取、观点分享的主要平台,在人们日常生活中占据重要地位。对社交网络信息进行挖掘,构建社交网络用户群体画像,对用户个性化服务、商业精准营销、网络舆情监控的引导有重要意义。当前,社交网络用户群体画像全面性、精准性有所欠缺,对此文章提出了一种基于社交网络数据的用户群体画像构建方法,对社交网络用户信息进行挖掘,从基本特征、内容特征、统计特征、行为特征等方面对用户群体进行全面精准刻画,充分了解用户群体兴趣偏好、行为倾向、价值诉求。使用机器学习、BP神经网络、LDA、特征融合等方法提取用户主题、表情符、发文习惯、位置等20类特征,构建社交网络用户个人画像,聚类分析得到用户群体,构建社交网络用户群体画像,为智能精准营销、平台个性化服务、舆情监控提供支撑。

With the development of network technology,social networks have become a major platform for people to obtain information and share opinions,which plays an important role in people's daily life.Mining social networks information and constructing social networks user group portraits is of great significance to user personalized services,commercial precision marketing,and online public opinion monitoring.Due to the lack of comprehensiveness and accuracy of current social network user group portrait,this paper proposes a social networks user group portrait method to mine the privacy information of social networks users,comprehensively and accurately depict user groups from basic features,content features,statistical features and behavior features to fully understand interest preferences,behavior tendencies and value demands of the user group.Using machine learning,BP neural network,LDA and feature fusion to extract 20 kinds of features such as user topics,emoticons,posting habits,and locations,to comprehensively characterize social networks users.Then use cluster analysis to obtain the user groups,and construct a complete social networks user group portrait to provide support for intelligent precision marketing,platform personalized service,public opinion monitoring.