《计算机应用研究》|Application Research of Computers

基于长短期记忆网络的社区演化预测

Prediction of community evolution based on long-short term memory network

免费全文下载 (已被下载 次)  
获取PDF全文
作者 蒋乐乐,刘厚泉,张楠
机构 中国矿业大学 计算机科学与技术学院,江苏 徐州 221116
统计 摘要被查看 次,已被下载
文章编号 1001-3695(2020)12-016-3599-04
DOI 10.19734/j.issn.1001-3695.2019.09.0543
摘要 现实生活中的网络通常是动态的,网络结构随着时间的推移而改变,检测社区演化可以深入了解网络的基本行为。针对动态社区演化预测问题,提出一种结合演化树和长短期记忆网络的社区演化预测方法,从动态网络中提取社区的多元特征,并使用长短期记忆网络对特征进行学习分类,最终预测社区下一时间段的变化情况。在两个真实数据集上进行了实验,实验结果证明该方法可以有效地预测社区演化行为,与其他方法相比具有较好的准确性。
关键词 动态网络; 社区演化预测; 长短期记忆网络
基金项目
本文URL http://www.arocmag.com/article/01-2020-12-016.html
英文标题 Prediction of community evolution based on long-short term memory network
作者英文名 Jiang Lele, Liu Houquan, Zhang Nan
机构英文名 School of Computer Science & Technology,China University of Mining & Technology,Xuzhou Jiangsu 221116,China
英文摘要 The network in real life is usually dynamic and the network structure evolves over time. Detecting the evolution of the community can understand the basic behavior of the network. To solve the problem of dynamic community evolution prediction, the paper proposed a community evolution prediction method based on evolutionary tree and long-short term memory network. Firstly, it extracted the multi-features of the community from the dynamic network. Then, it classified the features by using the long-short term memory network. Finally, it predicted the changes of the community in the next period of time. Experiments on two real data sets show that the proposed method can effectively predict community evolution behavior and is more competitive than other methods.
英文关键词 dynamic network; community evolution prediction; long-short term memory
参考文献 查看稿件参考文献
 
收稿日期 2019/9/24
修回日期 2019/11/19
页码 3599-3602,3617
中图分类号 TP301
文献标志码 A