摘要
情景演化分析系统的核心是利用大数据技术提取历史数据特征建立关联序列,通过深度学习建立无标记网络动力学模型,基于长短时记忆模型和动态网络方法对个体历史活动轨迹、网络行为、关联关系的演化进行分析,研究个体行为的规律,为个体行为分析预测提供有效的技术手段,解决个体行为随时间序列情景演化分析的实际问题。
Situational evolution analysis system is the core of big data technology is used to extract the characteristics of history data associated sequences,through deep learning unmarked network dynamics model is established,based on the length of the memory model and dynamic network method for individual course activity history,network behavior,the evolution of the correlation analysis,study the law of individual behavior,provide effective technical means for analysis and prediction of individual behavior,the solution to individual behavior over time sequence evolution analysis of practical problems.
作者
李道远
刘诚傲
黄昌金
曾青军
王庆友
卢翠平
吴刘青
LI Dao-yuan;LIU Cheng-ao;HUANG Jin-chang;ZENG Qing-jun;WANG Qing-you;LU Cui-ping;WU Liu-qing(Guangzhou Intelligence Communications Technology Co.,Ltd.Guangzhou 510000,China;China Academy of Electronic and Information Technology,Beijing 100041,China)
出处
《中国电子科学研究院学报》
北大核心
2020年第8期796-801,共6页
Journal of China Academy of Electronics and Information Technology
关键词
情景演化
长短时记忆
动态网络方法
situational evolution
long and short time memory
dynamic network method
