社交网络时空大数据聚类挖掘有效选择分析

来自社交网络的时空大数据具有海量和高动态的特性,有效选择时空数据进行聚焦挖掘分析至关重要.以微博位置签到数据为例,首先,对时空大数据空间聚类挖掘的有效选择问题进行了研究,针对社交网络时空数据不确定性问题,提出了时空大数据针对聚类挖掘的有效选择方法.聚类挖掘有效选择方法提出从空间、时间或属性等维度对时空大数据进行分割.然...

Effective Selecting Analysis About Cluster Mining for Social Media Big Data Analysis

Social media big data has the characteristics of a great quantity and high dynamic,which affects the accuracy and efficiency of the spatiotemporal data analysis.It is very important to select spatiotemporal data effectively from vas volume big data for the focusing mining analysis.Taking Weibo check-in data as an example,this paper studies the method of selecting effective datasets from spatiotemporal big data under cluster mining approach.According to the uncertainty of social media big data,an efficient selection(ES)method about cluster mining is proposed to select effective datasets.This method segments large data from the dimension of space,time or attribute.Then do exploratory spatial data analysis(ESDA)to the subsets to get the datasets with the potentiality of cluster mining.Finally,the proposed method is verified by using the Weibo check-in data in Wuhan which is between 2011and 2015to mine commercial hotspots.The results show that ES can extract small dataset from social media big data,which can get the same result as the original big data in data mining.And the smaller sometimes even can increase the efficiency of spatiotemporal big data mining,and the mining accuracy maybe higher.