Fraud phone number identification method and system

Abstract

Embodiments of the present invention provide a method and system for identifying fraudulent phone numbers. The identification method includes: performing feature extraction on the call record, and obtaining the call record after the feature extraction. The call record after the feature extraction includes the proportion of empty numbers and the proportion of calls by friends and the entropy of the phone number in unit time; the call record after the feature extraction is input into the trained fraud call recognition model, and the fraud call recognition result corresponding to the call record is obtained, wherein the fraud call recognition model passes the gradient Boost decision tree algorithm training. The method and system for identifying fraudulent phone numbers provided by the embodiments of the present invention can perform more detailed and accurate identification of whether a phone number is a fraudulent phone number; by setting a training model through a gradient-raising decision tree algorithm, it is possible to identify fraudulent phone numbers more accurately .

Description

technical field [0001] The embodiments of the present invention relate to the field of communication network security, and more specifically, to a method and system for identifying fraudulent phone numbers. Background technique [0002] In recent years, the practice of using telephones to commit fraud and disseminate information such as violence, terror, and cults has exploded. Bad calls have become a pain point for the general public and a chronic social disease. In 2013, the amount of defrauded cases caused by bad calls reached 10 billion yuan; in 2014, the number of cases caused by bad calls increased greatly on the basis of the previous year; The defrauded amount has doubled on the basis of the previous year, and various cases caused by bad calls are shocking. Behind these losses is the black industry of communication information fraud. The entire black industry has formed an industrial chain with clear division of labor and efficient operation. From the perspective of ...

AI Technical Summary

Problems solved by technology

Bad calls not only caused the smog of the communication network and the deterioration of the ecology, but also caused huge property losses and even life safety to the people

[0004] At present, the acquisition and identification of fraudulent calls is mainly through voice recognition. This method is not very learning, and the acquisition and identification are not accurate enough.

Images