Method and device for identifying non-real person incoming calls

Abstract

The embodiment of this specification provides a non-real person incoming call identification method and device, the method includes: the called party's intelligent gateway device receives the calling party's incoming call; sends a first interactive answer to the calling party with a knowledge question Asking questions; receiving a first answer of the first interactive response question from the calling party; when it is determined that the first answer is correct, sending a second interactive answer question of reasoning questions to the calling party; Receive a second answer to the second interactive response question from the calling party; when it is determined that the second answer is correct, identify that the incoming call of the calling party is a real-person incoming call; or, when it is determined that the When the second answer is wrong, it is recognized that the incoming call of the calling party is a non-human incoming call. It can realize the pre-prevention of telecommunications fraud, efficiently intercept fraudulent calls, and does not affect the privacy of communication.

Description

technical field [0001] One or more embodiments of this specification relate to the field of intelligent control, and in particular, to a method and device for identifying non-human incoming calls. Background technique [0002] In the early stage of telecommunications fraud, the method of "spreading the net" is generally used to spread a large amount of fraud bait information to lure the victim to contact the fraudster. A typical feature of telecommunications fraud is non-real person calls. It is a relatively elementary form to play fixed recordings on the phone. At the high end, speech recognition technology can be used to design interactive dialogue strategies to accurately screen fraud targets and match them with appropriate fraud scripts. Usually, interested customers will transfer the call to a special seat for further brainwashing and deception through role-playing. It can be seen that it is a better solution to identify fraudulent calls in the initial stage of call ac...

AI Technical Summary

Problems solved by technology

Therefore, most of the current anti-fraud methods are inferred through the interception and analysis of the phone content, which affects the privacy of communication.

There are also voiceprint features of high-frequency fraudulent calls to find matching calls. Usually, voiceprint information needs to be obtained in advance, and the efficiency of real-time interception is doubtful.

Or carry out after-the-fact reminders and precautions for high-risk incoming calls (such as overseas calls), which cannot be prevented in advance

Images