Design And Implementation Of Mobile CDR Anomaly Discovery And Location System Based On Statistical Analysis

As one of the core competitive content of the communication industry,the accuracy of communication billing billing has been increasingly concerned by various communication operators.To ensure the accuracy of billing,it is necessary to start from the verification of the batch result,the monitoring and analysis of the abnormal bills,and the data balance comparison of each link.Through multi-level,multi-dimensional,multi-view scan analysis and model check identification and mining technology.To achieve an overall improvement in billing quality.Therefore,the monitoring and analysis of abnormal bills is an important part of ensuring billing accuracy,and plays an important role in discovering system vulnerabilities,preventing revenue loss,and avoiding customer complaint risks.In the design process,the system uses WEB architecture technology,uses eclipse tools for development,the database uses Oracle design,and the whole system uses WebLogic for deployment and management.Through the analysis and research of all abnormal bill characteristics,the four analysis dimensions covering the whole abnormal bill category are established,namely,the service type dimension,the billing link dimension,the abnormal category dimension and the bill source dimension,and use this as the entry point.The statistical model of the overall data of abnormal bills is constructed.Based on the overall analysis and the presentation of the problems,the causes of the problems are finally determined and the subsequent solutions are taken to realize the abnormal bills of the network cutting that cannot be realized in the previous manual analysis.The analysis function fills the gap in the analysis of network cutting abnormal bills.The system developed this time mainly uses multi-dimensional and multi-level scientific statistical analysis to automatically find and locate the problem of network elements or systems from the statistical data of abnormal bills.Analyze the model based on the scientific analysis of abnormal vocabulary features,so that the system can automatically construct statistical actions,and explore the problems reflected by the data from the statistical data.The original manual abnormal bill analysis mode is changed to the computer automatic analysis mode,and the abnormal problem is highlighted to the analyst with a clear and simple human-machine interface,thereby improving the analysis ability of the analyst.The daily analysis of the networkabnormal bills is realized,which fills the blank of the abnormal bill analysis in our province,and constructs an analysis method that gradually extends from the problem to the detail,and normalizes the abnormal bill analysis.