Design And Implementation Of Telecom Fraud Recognition Classifier Based On Convolutional Neural Network

With the advent of mobile internet and the era of big data,the sharing and exchange of information brings new value to people,and at the same time,people's personal information is easily used by criminals and even used for telecom fraud.Systematic analysis of telecommunication subscriber numbers through artificial intelligence and other means can effectively identify telecom fraudulent subscriber numbers,detect telecom fraud early,and impose some restrictions on these numbers to kill telecom fraud in the cradle.Aiming at the above requirements,this paper designs a telecommunication scam recognition classifier system based on convolutional neural network.After the user CDR data is cleaned and processed into a picture,the convolutional neural network is used to construct the classifier,and the telecom user CDR is Data learning,mining fraud user call data characteristics,identifying fraud user behavior.Firstly,the paper introduces the relevant background and technology,and then analyzes the overall requirements of the telecom fraud recognition classifier based on convolutional neural network,and decomposes each specific functional requirement.Then the overall module architecture of the system is designed,and the interaction design and data structure definition of each module are given.On this basis,the detailed design scheme of the system is introduced.The specific attributes and methods of each class are described through the class diagram,and the related relationships are given.The detailed key processes and designs are given.Then the test cases were designed,all the functional modules were tested and some screenshots were shown.Finally,the design and implementation of the system are summarized and the future research directions and optimization directions are pointed out.