一种基于遗传算法的网络多源数据流共享查询方法  

A shared query method for network multi-source data streams based on genetic algorithm

作　　者：牟凯 赵可英 MU Kai;ZHAO Ke-ying(Practice Teaching Center(Information Center),Sichuan Business Vocational College,Chengdu,Sichuan 611131,China;School of Mathematics and Information Sciences,Neijiang Normal University,Neijiang,Sichuan 641100,China)

机构地区：[1]四川商务职业学院实践教学中心(信息中心),四川成都611131 [2]内江师范学院数学与信息科学学院,四川内江641100

出　　处：《宁德师范学院学报(自然科学版)》2024年第4期367-373,共7页Journal of Ningde Normal University(Natural Science)

摘　　要：针对传统流查询均为集中流查询问题,无法分配任务至多个节点,存在单点瓶颈,难以准确、快速进行共享查询等问题,基于遗传算法设计网络多源数据流共享查询方法。通过网络数据流管理系统多源查询数据流,再通过遗传算法编码选择最优查询路径,最后根据投影操作符共享网络数据流,从而高效完成对网络多源数据流的共享查询。实验结果表明,在新增数据量为5000个时,文中方法查询时间仅为400 s,曲线下面积(AUC)值为0.9852,相较其他方法查询时间最短,准确性较高,能够准确、快速地对网络多源数据流实现共享查询,应用效果较好。Due to the fact that traditional flow queries are all centralized flow queries,tasks cannot be assigned to multiple nodes,resulting in a single point bottleneck problem that makes it difficult to accurately and quickly perform shared queries.Therefore,a network multi-source data flow shared query method based on ge⁃netic algorithm is studied.By using a network data flow management system to query data streams from mul⁃tiple sources,genetic algorithm encoding is used to select the optimal query path,and finally,network data streams are shared based on projection operators to efficiently complete shared queries for network multisource data streams.The experimental results show that when the amount of newly added data is 5000,the query time of our method is only 400 s,with an area uder the curve(AUC)value of 0.9852,which is the short⁃est and has high accuracy.It can accurately and quickly share and query multi-source data streams in the net⁃work,and the application effect is good.

关 键 词：遗传算法 数据流共享 查询优化 网络多源数据流 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]