CN-115952336-B - Identification method and system for impersonation user information

CN115952336BCN 115952336 BCN115952336 BCN 115952336BCN-115952336-B

Abstract

The invention discloses a method and a system for identifying impersonation user information, and relates to the technical field of information identification. The method comprises the steps of obtaining suspected user information, preprocessing the suspected user information to obtain suspected user data, further matching the suspected user data with all user data to be protected in a preset protection library, sorting all the user data to be protected in a descending order according to a matching result to obtain user sorting data to be protected, finally filtering the user data to be protected based on a preset recall strategy and the user sorting data to be protected, marking the user data to be protected meeting preset requirements as filtered user data, outputting the user information to be protected corresponding to the filtered user data as impersonated user information, and effectively solving the technical problem that the identification efficiency is low when the suspected user information (impersonated nickname and/or head portrait) is identified in the prior art.

Inventors

ZHANG ZHENGTONG
WANG WEIZHE
PAN ZISHENG
HUANG XIANGKANG
ZENG RUIHONG
LAN XIANG
XIONG JIA
XU ZHIJIAN
XIE RUI
CHEN GUANGYAO
LI ZIJUN
MA JINLONG
YANG HANYUE
Yuan Runa
WU WENLIANG
YE XINHUA
WU HUIYANG
DENG QICHUN

Assignees

广州趣丸网络科技有限公司

Dates

Publication Date: 20260508
Application Date: 20221229

Claims (8)

1. A method of identifying impersonation of user information, comprising: s1, obtaining information of a suspected user, wherein the information of the suspected user comprises a suspected nickname and a suspected head portrait, and the suspected nickname and the suspected head portrait are matched; S2, preprocessing the suspected user information to obtain suspected user data, wherein the preprocessing of the text of the suspected nickname to obtain suspected nickname data; S3, matching the suspected user data with all user data to be protected in a preset protection library based on a preset matching algorithm, and sorting all the user data to be protected according to a matching result to obtain sorting data of the users to be protected, wherein the user data to be protected comprises nickname data to be protected and head portrait data to be protected; S3C, matching the suspected nickname data with all nickname data to be protected in a preset protection library based on the preset nickname matching algorithm, sorting all nickname data to be protected according to nickname matching results to obtain nickname sorting data to be protected, filtering the nickname data to be protected based on a preset nickname recall strategy and the nickname sorting data to be protected, marking the nickname data to be protected meeting preset nickname requirements as filtering nickname data, marking the nickname data to be protected matched with the filtering nickname data as first head image data to be protected, matching the suspected head image data with the first head image data to be protected based on the preset head image matching algorithm, and sorting all the first head image data to be protected according to the first head image matching results to obtain first head image sorting data to be protected; S4, filtering the user data to be protected based on a preset recall strategy and the user ordering data to be protected, marking the user data to be protected meeting preset requirements as filtered user data, and outputting user information to be protected corresponding to the filtered user data as impersonated user information; S4C, filtering the first head image data to be protected based on the preset nickname head image joint recall strategy and the first head image ordering data, recording the first head image data to be protected meeting the preset nickname head image joint requirement as first filter head image data, outputting the head image to be protected corresponding to the first filter head image data as an impostor head image, and outputting the name to be protected corresponding to the nickname data to be protected matched with the first filter head image data as an impostor nickname.
2. The method for identifying impersonation of user information according to claim 1, wherein, S3D, based on the preset head portrait matching algorithm, matching the suspected head portrait data with all head portrait data to be protected in a preset protection library, and sorting all the head portrait data to be protected in a descending order according to head portrait matching results to obtain head portrait sorting data to be protected; Filtering the head portrait data to be protected based on the preset head portrait recall strategy and the head portrait sequencing data to be protected, marking the head portrait data to be protected meeting the requirement of the preset head portrait as filter head portrait data, and marking the nickname data to be protected, which are matched with the filter head portrait data, as first nickname data to be protected; Based on the preset nickname matching algorithm, matching the suspected nickname data with the first nickname data to be protected, and sorting all the first nickname data to be protected according to a first nickname matching result to obtain first nickname sorting data to be protected; after step S3D, step S4 is specifically: S4D, filtering the first nickname data to be protected based on the preset head portrait nickname joint recall strategy and the first nickname ordering data, marking the first nickname data to be protected meeting the preset head portrait nickname joint requirement as first filtering nickname data, outputting the nickname to be protected corresponding to the first filtering nickname data as a masquerade nickname, and outputting the head portrait to be protected corresponding to the head portrait data to be protected matched with the first filtering nickname data as a masquerade nickname.
3. The method for identifying masquerading user information according to claim 1, wherein performing text preprocessing on the suspected nickname to obtain suspected nickname data comprises: Carrying out numerical normalization, letter case conversion, chinese character simplified conversion, mars conversion and general word filtering on the suspected nickname to obtain suspected nickname data; Or vector conversion is carried out on the suspected nickname based on a preset semantic characterization model, so as to obtain suspected nickname data.
4. The method for identifying impersonation user information according to claim 1, wherein the image preprocessing is performed on the suspected head portrait to obtain suspected head portrait data, and the method specifically comprises the following steps: Performing image scaling, image overturning, image binarization and image compression on the suspected head portrait to obtain suspected head portrait data; or vector conversion is carried out on the suspected head portrait based on a preset image characterization model, so as to obtain suspected head portrait data.
5. The method for identifying impersonation of user information according to claim 1, further comprising, after step S4: And storing the impersonated user information as updated protection user information, and updating the preset matching algorithm based on the updated protection user information.
6. The method for identifying impersonation of user information according to claim 1, further comprising, after step S4: and packaging, storing and displaying the faked user information and the corresponding suspected user information.
7. The method for identifying impersonation of user information according to claim 1, further comprising, after step S4: and sending a masquerade risk reminder to the masquerade user corresponding to the masquerade user information.
8. A system for identifying impersonation of user information, comprising: The device comprises a suspicion information acquisition module, a suspicion information preprocessing module and a suspicion information processing module, wherein the suspicion information acquisition module acquires suspicion user information, the suspicion user information comprises a suspicion nickname and a suspicion head image, the suspicion nickname is matched with the suspicion head image, and the suspicion information preprocessing module is used for preprocessing the suspicion user information to obtain suspicion user data, and comprises the steps of conducting text preprocessing on the suspicion nickname to obtain suspicion nickname data, or conducting image preprocessing on the suspicion head image to obtain suspicion head image data; The information matching and sorting module is used for matching the suspected user data with all user data to be protected in a preset protection library based on a preset matching algorithm, and sorting all the user data to be protected in a descending order according to a matching result to obtain user sorting data to be protected, wherein the user data to be protected comprises nickname data to be protected and head portrait data to be protected; the information matching and sorting module is used for executing S3C, based on the preset nickname matching algorithm, matching the suspected nickname data with all nickname data to be protected in a preset protection library, sorting all nickname data to be protected according to nickname matching results to obtain nickname sorting data to be protected, filtering the nickname data to be protected based on a preset nickname recall strategy and the nickname sorting data to be protected, marking the nickname data to be protected meeting the requirement of the preset nickname as filtering nickname data, marking the head image data to be protected matched with the filtering nickname data as first head image data to be protected, and based on the preset head image matching algorithm, matching the suspected head image data with the first head image data to be protected, and sorting all the first head image data to be protected according to the first head image matching results to obtain first head image sorting data to be protected; the identification output module is used for filtering the user data to be protected based on a preset recall strategy and the user ordering data to be protected, recording the user data to be protected meeting preset requirements as filtered user data, and outputting the user information to be protected corresponding to the filtered user data as impersonated user information; When the information matching and sorting module is used for executing the S3C, the identification output module is used for filtering the first head image data to be protected based on the preset nickname head image joint recall strategy and the first head image sorting data, marking the first head image data to be protected meeting the preset nickname head image joint requirement as first filter head image data, outputting the head image to be protected corresponding to the first filter head image data as an imposter head image, and outputting the nickname to be protected corresponding to the nickname data to be protected, which is matched with the first filter head image data, as an imposter nickname.

Description

Identification method and system for impersonation user information Technical Field The invention relates to the technical field of information identification, in particular to a method and a system for identifying impersonated user information, in particular to a method and a system for identifying impersonated nicknames and/or head portraits. Background When a user uses social software, the user can have a proprietary head portrait and a nickname, and the head portrait and the nickname become important bases for judging the identity of the user. At the same time, spoofing of the head and/or nickname of the masquerading user also occurs. The user to be made is fraudulent by impersonating the head portraits and/or nicknames of the user to be protected, and under the scene, the impersonated nicknames and/or head portraits can be effectively identified by calculating the similarity of the head portraits and/or nicknames, so that the fraudulent behavior is identified. The existing image/audio/text/video and other media data can be subjected to similarity calculation after characterization, the similarity between the media data can be intuitively quantized, and different quantization methods can generate different results. In an actual anti-impersonation fraud business scene, whether the suspected nickname is an impersonation nickname is generally judged through matching search, and whether the suspected nickname is an impersonation nickname is judged through manually comparing the suspected nickname with the head images in the protection library. Although the above approach can solve the problem of complete impersonation of some avatars and/or nicknames, the recognition efficiency is low, which is not beneficial to timely finding impersonation/fraud. Disclosure of Invention The invention provides a method and a system for identifying impersonation user information, which are used for solving the technical problem of low identification efficiency in the prior art when suspicious user information (impersonation nickname and/or head portrait) is identified. The invention provides a method for identifying impersonation user information, which comprises the following steps: S1, obtaining suspected user information; s2, preprocessing the suspected user information to obtain suspected user data; S3, matching the suspected user data with all user data to be protected in a preset protection library based on a preset matching algorithm, and sorting all the user data to be protected according to a matching result to obtain sorting data of the users to be protected; And S4, filtering the user data to be protected based on a preset recall strategy and the user ordering data to be protected, marking the user data to be protected meeting preset requirements as filtered user data, and outputting the user information to be protected corresponding to the filtered user data as impersonated user information. Preferably, the method comprises the steps of, The suspected user information comprises a suspected nickname and a suspected head image, wherein the suspected nickname is matched with the suspected head image; The user data to be protected comprises nickname data to be protected and head portrait data to be protected, wherein the nickname data to be protected and the head portrait data to be protected are matched; the preset matching algorithm is a preset nickname matching algorithm or a preset head portrait matching algorithm; The preset recall strategy is a preset nickname recall strategy or a preset head portrait recall strategy or a preset nickname head portrait joint recall strategy or a preset head portrait nickname joint recall strategy; The preset requirement is a preset nickname requirement or a preset head portrait requirement or a preset nickname head portrait combination requirement or a preset head portrait nickname combination requirement; the filtering user data is filtering nickname data or filtering head portrait data or filtering nickname head portrait joint data or filtering head portrait nickname joint data. Preferably, step S2 includes: S2A, performing text preprocessing on the suspected nickname to obtain suspected nickname data. Preferably, step S2 further includes: S2B, performing image preprocessing on the suspected head portrait to obtain suspected head portrait data. Preferably, step S3 includes: and S3A, based on the preset nickname matching algorithm, matching the suspected nickname data with all nickname data to be protected in a preset protection library, and sorting all nickname data to be protected according to nickname matching results to obtain nickname sorting data to be protected. Preferably, step S3 includes: And S3B, based on the preset head portrait matching algorithm, matching the suspected head portrait data with all head portrait data to be protected in a preset protection library, and sorting all the head portrait data to be protected according to head portrait matching results to o