Runan Song, Yang Xue, P. Zhang, Yining Yang, Cong Wang
{"title":"Multi-label co-occurrence network for stealing electricity type detection and research on forensics sequence rules","authors":"Runan Song, Yang Xue, P. Zhang, Yining Yang, Cong Wang","doi":"10.1145/3579654.3579763","DOIUrl":null,"url":null,"abstract":"With the development of economy and the progress of technology, the anti-stealing and investigation work in power consumption management has become extremely complex and arduous. In the complex environment of the electricity stealing scene, due to the hidden electricity stealing and the complex on-site power distribution relations, it is difficult to carry out the on-site evidence collection work smoothly. This paper focuses on the stealing electricity type detection and the process of on-site evidence collection, deeply analyzes the key elements of electricity stealing, and constructs a stealing electricity type detection model based on multi label graph convolution network. It uses GRU network to mine the time sequence characteristics of power consumption, learns the co-occurrence relationship of electricity stealing labels based on GCN network, and combines the user attribute characteristics to ultimately improve the accuracy of stealing electricity type detection. Based on the above, by analyzing the key elements such as the historically collected data, we use the method of combining expert experience and statistical analysis to generate the forensics sequence rules of evidence collection for different types of electricity stealing, thus effectively improving the efficiency and quality of evidence collection for on-site staff. The experimental results demonstrate that the proposed methods perform favorably against the most frequently used methods.","PeriodicalId":146783,"journal":{"name":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","volume":"83 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings of the 2022 5th International Conference on Algorithms, Computing and Artificial Intelligence","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1145/3579654.3579763","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
With the development of economy and the progress of technology, the anti-stealing and investigation work in power consumption management has become extremely complex and arduous. In the complex environment of the electricity stealing scene, due to the hidden electricity stealing and the complex on-site power distribution relations, it is difficult to carry out the on-site evidence collection work smoothly. This paper focuses on the stealing electricity type detection and the process of on-site evidence collection, deeply analyzes the key elements of electricity stealing, and constructs a stealing electricity type detection model based on multi label graph convolution network. It uses GRU network to mine the time sequence characteristics of power consumption, learns the co-occurrence relationship of electricity stealing labels based on GCN network, and combines the user attribute characteristics to ultimately improve the accuracy of stealing electricity type detection. Based on the above, by analyzing the key elements such as the historically collected data, we use the method of combining expert experience and statistical analysis to generate the forensics sequence rules of evidence collection for different types of electricity stealing, thus effectively improving the efficiency and quality of evidence collection for on-site staff. The experimental results demonstrate that the proposed methods perform favorably against the most frequently used methods.