Fraudulent telephone number analysis method and equipment

Abstract

The invention provides a fraudulent telephone number analysis method and equipment. The analysis method comprises the following steps: acquiring abnormal behavior data and feature data of telephone numbers in an original telephone bill, wherein the abnormal behavior data include one or more of number of calls of abnormal numbers, number of calls of vacant numbers, and number of calls of strange numbers, and the feature data comprise activity degree and call data, wherein the activity degree comprises the daily call frequency and / or a call intensive time period, and the call data comprise one or more of call completing rate, earliest call time, latest call time, average call duration, main call area and average ring time; and inputting two kinds of data into a trained fraudulent telephone number analysis model, and acquiring a fraudulent telephone number analysis result through a weighted naive Bayesian classification algorithm. The fraudulent telephone number analysis method provided by the invention can accurately analyze whether a telephone number is a fraudulent telephone number, and a multi-dimensional fraudulent telephone number analysis model is obtained, so that whether thetelephone number is the fraudulent telephone number can be comprehensively analyzed.

Description

technical field [0001] The invention relates to the field of telecommunications fraud monitoring and protection, and more specifically, to a method and device for analyzing fraudulent telephone numbers. Background technique [0002] In recent years, there has been an explosion of fraud incidents using telephone calls, with a wide range of victims and huge sums of money. Telecom fraud has become a pain point for telecom users. There are also various means of fraudulent phone calls at present. Scammers often change their numbers through software, and impersonate banks, operators, acquaintances, social security, etc. to carry out fraudulent activities. In 2013, there were more than 300,000 communication fraud cases across the country, and the victims were defrauded of nearly 10 billion yuan; in 2014, there were more than 400,000 communication fraud cases nationwide, and the victims were defrauded of nearly 10.7 billion yuan; There were 590,000 cases, a year-on-year increase of...

AI Technical Summary

Problems solved by technology

In the current telecommunications fraud, scammers have a variety of fraudulent means, which is hard to guard against, which seriously affects the security of telecommunications and the order of calls, and damages the interests of telecommunications users and the reputation of telecommunications networks.

[0004] In view of the frequent occurrence of the above-mentioned telecommunication fraud cases and the bad social influence, some methods for analyzing and comparing fraudulent telephone numbers have appeared. Secondary screening and comparison, and a series of complex calculations, these methods are more complicated, not simple and comprehensive enough

Images