基于云边协同的社交网络大数据聚类挖掘方法  

A Clustering Mining Approach for Social Network Big Data Based on Cloud⁃Edge Collaboration

作　　者：胡小琴 HU Xiaoqin(Software Institute,Quanzhou University of Information Engineering,Quanzhou Fujian 362000,China)

机构地区：[1]泉州信息工程学院软件学院,福建泉州362000

出　　处：《海南热带海洋学院学报》2024年第5期67-72,87,共7页Journal of Hainan Tropical Ocean University

基　　金：福建省教育厅中青年教师教育科研项目(JAT_(1)90930)。

摘　　要：社交网络数据量庞大,若缺乏高带宽和低延迟技术支持,将导致聚类挖掘尺度误差的增大和精度的降低。为解决以上问题,本文提出了一种基于云边协同的社交网络大数据聚类挖掘方法。首先,利用云边协同计算的能力,对社交网络大数据进行挖掘处理,并重组其空间区域结构信息。其次,基于云边协同计算的特征统计能力,进行特征信息逻辑推理,对挖掘得到的社交网络大数据进行分析。最后,在云边协同空间区域内,重新组织社交网络特征数据的聚类结构,从而更准确地确定聚类中心,实现对社交网络大数据的聚类挖掘。通过实验证明,该方法有效降低了挖掘偏差,提高了挖掘过程中特征尺度推进能力,并优化了挖掘环境。When lacking high bandwidth and low latency technical support,the huge amount of data in social networks will lead to an increase in clustering mining scale and a decrease in accuracy.To address the above problems,a clustering mining method for social network big data based on cloud‐edge collaboration was proposed.Firstly,cloud‐edge collaborative computing was utilized to mine and process social network big data and reorganize its spatial regional structure information.Secondly,based on the feature statistical capability of cloud‐edge collaborative computing,logical inference of feature information was carried out to analyze the mined social network big data.Finally,within the cloud‐edge collaborative spatial region,social network feature data’s clustering structure was reorganized to determine the clustering center more accurately and achieve clustering mining of social network big data.The results showed that the present method effectively reduces the mining bias,improves the ability to advance feature scales during the mining process,and optimizes the mining environment.

关 键 词：云边协同 社交网络 大数据 聚类挖掘 低延迟技术 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]