基于深度学习的IT服务综合监控系统异构数据集成方法  

Heterogeneous Data Integration Method of IT Service Integrated Monitoring System Based on Deep Learning

作　　者：杨航 卢伟开 黄海英 吕华辉 魏理豪 YANG Hang;LU Weikai;HUANG Haiying;L Huahui;WEI Lihao(Digital Grid Research Institute,China Southern Power Grid,Guangzhou 510663,China)

机构地区：[1]南方电网数字电网研究院有限公司,广东广州510663

出　　处：《微型电脑应用》2023年第3期68-70,共3页Microcomputer Applications

摘　　要：在IT服务综合监控系统数据集成处理过程中,由于受到数据检测技术的影响,使得数据集成F-Score值较低,无法满足异构数据集成需求。因此,提出基于深度学习的IT服务综合监控系统异构数据集成方法。针对IT服务综合监控系统中的异构数据,通过数据清洗和数据转换进行预处理。依托于DDE技术,将处理后的数据传输至服务器端。利用半监督深度学习方法,构建数据检测模型,完成异构数据的属性检测。在完成异构数据特征关联度约束后,实现数据的高精度集成处理。仿真实验结果表明,应用所提方法能够有效提升数据集成处理能力。In the data integration process of the IT service integrated monitoring system,the data detection technology is affected,which makes the data integration F-Score value low,and cannot meet the needs of heterogeneous data integration.Therefore,a heterogeneous data integration method based on deep learning of IT service integrated monitoring system is proposed.For the heterogeneous data in the IT service integrated monitoring system,pre-processing is performed through data cleaning and data conversion.Relying on DDE technology,the processed data are transmitted to the server.The semi-supervised deep learning method is used to construct a data detection model to complete the attribute detection of heterogeneous data.After completing the constraint of the feature relevance of heterogeneous data,high-precision integrated processing of data is realized.The simulation experiment results show that the application of the proposed method can effectively improve the data integration processing capability.

关 键 词：深度学习 IT服务 综合监控系统 异构数据 数据传输 

分 类 号：TM72[电气工程—电力系统及自动化]