An information identification method and server

Abstract

The invention discloses an information identification method and a server, comprising: obtaining first information of a first user, the first information representing the user credit; obtaining second information of the first user, the second information representing the user Basic information: obtain the similarity value between the first user and each second user according to the second information of the first user and the second information of more than one second user, and the difference between the more than one second user and the target user There is an association relationship between them; when the first information of the first user satisfies the first condition, and there is a similarity value that satisfies the second condition among the similarity values ​​between the first user and each second user, determine the The identity of the first user is a forged identity; and a prompt message is sent to the target user.

Description

technical field [0001] The invention relates to information identification technology, in particular to an information identification method and server. Background technique [0002] Social networks make social interaction between users faster and more convenient. At the same time, some users use forged identities to commit fraud on social networks. In order to remind users to beware of fraud, keyword reminders and behavioral fingerprint recognition are usually used. Keyword reminder: When the chat record involves related keywords such as borrowing money, the chat tool reminds the user to pay attention to the identity of the other party in the form of text in the chat conversation to prevent fraud. Behavioral fingerprinting: Build a user identification model based on the user's habit of using software (such as keyboard input speed, etc.), to distinguish whether the same account belongs to the same user. [0003] In the prior art, fraudsters can easily bypass the fraud remi...

AI Technical Summary

Problems solved by technology

[0003] Existing technology is easy for scammers to avoid fraud reminder checks by repeatedly trying to avoid keywords

Moreover, some scammers currently forge their identities by forging friends' avatars and nicknames to confuse users, which reduces the effect of prompts triggered by keywords

On the other hand, when fraudsters forge other people's identities and apply for friends, the chat has not yet started, the keyword reminder has not been triggered, and the user cannot be warned in time

Behavioral fingerprinting identifies the situation that the same friend account is stolen, etc., and the behavior difference generated by the second user cannot identify the fraudulent behavior generated by another account by forging friend information.

Images