高级检索

    离线瞬态社会网络中的多用户位置邻近预测

    Multi-User Location Proximity Prediction in Offline Ephemeral Social Networks

    • 摘要: 离线瞬态社会网络(offline ephemeral social network, OffESN)是一种在特定时间通过参加特定事件临时组建的新型社会网络.随着移动智能终端的普及和短距离通信技术(如蓝牙、RFID技术等)的发展,该类型网络得到工业界和学术界越来越多的关注.位置邻近(location proximity)关系是指用户在离线网络中的相遇关系.针对位置邻近关系的动态变化性及持续时间短等特征,主要研究离线瞬态社会网络中多用户邻近关系预测问题.首先,给出离线瞬态社会网络中的相关概念及问题定义;然后,设计多用户邻近关系预测总体框架,包括网络片段收集、叠加网络构建、网络过滤及极大紧密子图发现等步骤.由于多邻近关系的数量及每个邻近关系中用户的数量不能事先确定,基于分裂思想提出了一种极大紧密子图挖掘策略,以预测多用户位置邻近关系.该挖掘算法是以加权边介数为网络分裂依据,以聚集密度为分裂结束条件.在2个真实数据集上完成了实验,验证了所提出预测策略的可行性及效率.

       

      Abstract: Offline ephemeral social network (OffESN) is defined as a new kind of offline social networks created at a specific location for a specific purpose temporally, and lasting for a short period of time. With the popularity of mobile intelligent terminals and the development of short distance communication technologies such as Bluetooth and RFID, the OffESN is receiving more and more attentions from industry and academic communities. Location proximity relations are encounter relations of the users in the OffESN. Aiming to the characteristics such as dynamic change and short duration time, this paper intends to study the problem of multi-user location proximity in the OffESN. First of all, the paper puts forward relevant concepts in the OffESN and defines the problem to be solved. Then, it designs the overall framework of multi-user location proximity prediction, including network segments collection, overlay networks construction, network filter and maximal close subgraphs discovery. Based on the framework and the splitting idea, the paper suggests a maximal close subgraph discovery algorithm for predicting multi-user location proximity. The algorithm uses weighted edge betweenness (WEB) as the basis of splitting, and uses the aggregate density as the termination condition of spitting, which can effectively solve the problem that both numbers of location proximity relations and the users in each location proximity are uncertain. Finally, the experiments on two real datasets verify the feasibility and efficiency of the suggested prediction strategy.

       

    /

    返回文章
    返回