In recent years

communication fraud has sustained serious

causing great concern from all walks of life. On the other side

with the development of 5 G technology

the proliferation of connected devices and sensors has led to an exponential increase in the number and speed of data streams

which also significantly increases the complexity of data processing for identifying and preventing fraud. Therefore

for telecom operators

traditional anti-fraud technology will no longer apply

and need to process massive amounts of data in a short period of time to prevent fraud. In order to solve the above problems

this paper proposes a method for managing communication information fraud. It can realize the extraction of the suspect number’s communication characteristics from the signaling

and quickly judge and dispose of it with machine learning method under 5 G. The accuracy of this method is up to 85% by local verification.