基于数据挖掘的退库信息异常告警方法  

Normal alarm method of database database information based on data mining

作　　者：赵夏 殷大新 刘睿 ZHAO Xia;YIN Daxin;LIU Rui(State Grid Jiuquan Power Supply Company,Jiuquan 735000,China)

机构地区：[1]国网酒泉供电公司,甘肃酒泉735000

出　　处：《无线互联科技》2024年第13期107-110,共4页Wireless Internet Technology

摘　　要：鉴于目前退库信息异常告警方法存在的误警率较高问题,文章提出一种基于数据挖掘的退库信息异常告警方法。首先,利用网络爬虫技术爬取退库信息,并对无效信息进行识别和清除。其次,对退库信息进行标准化处理,将需要变化的数据项从原有空间转换至统一的标准数据空间。再次,运用数据挖掘技术提取退库信息的异常特征,并运用赋权法对这些异常特征进行融合,以量化退库信息的异常程度及告警等级。最后,根据量化结果作出相应的告警行为,从而实现基于数据挖掘的退库信息异常告警。实验结果表明,该方法的设计误警率不超过1%,在退库信息异常告警领域具有广阔的应用前景。In view of the high false alarm rate of the current abnormal alarm method,this paper proposes a new abnormal alarm method based on data mining.First,the web crawler technology is used to crawl the database information,and the invalid information is identified and cleared.Then,the database return information is standardized,and the data items that need to be changed from the original space to the unified standard data space.Then,the data mining technology is used to extract the abnormal characteristics of the database information,and the empowerment method is used to integrate these abnormal characteristics to quantify the abnormal degree and alarm level of the database information.Finally,according to the quantitative results of the corresponding alarm behavior,so as to realize the abnormal alarm based on data mining.The experimental results show that the design false alarm rate of this method does not exceed 1%,and it has a broad application prospect in the field of abnormal alarm of database return information.

关 键 词：数据挖掘 退库信息 异常 告警 网络爬虫技术 赋权法 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]