Account data processing method and device, abnormal account identification method and device and computer equipment

Abstract

The invention relates to an account data processing method, an abnormal account identification method and device and computer equipment. The method comprises the following steps: acquiring changed head portraits corresponding to head portrait change events in an interactive media platform according to a preset time period; clustering the changed head portraits corresponding to the head portrait change events to obtain a head portrait clustering result; the head portrait clustering result comprises cluster information of at least one head portrait clustering cluster; the similarity between different changed head portraits belonging to the same head portrait cluster is greater than a preset similarity threshold; screening out abnormal head portrait clusters from the head portrait clusters; the number of the changed head portraits in the abnormal head portrait cluster is greater than a preset number threshold; extracting abnormal head portrait features of the abnormal head portrait cluster; the abnormal head portrait feature is used for identifying an abnormal user account in the interactive media platform. By adopting the method, the abnormal user account on the interactive media platform can be accurately identified.

Description

Technical Field

[0001] The present application relates to the field of big data technology, and in particular to an account data processing method, an abnormal account identification method, an apparatus, a computer device, a computer-readable storage medium, and a computer program product. Background Art

[0002] In current social applications, abnormal users change their profile pictures in batches within a short period of time and use various channels of social applications to increase their exposure, thereby achieving improper purposes.

[0003] At present, when identifying abnormal user accounts in social applications, cluster analysis is usually performed on business traffic data to identify abnormal traffic. However, although this method of clustering business traffic data can effectively identify abnormal traffic, when it is applied to the analysis of abnormal user accounts changing avatars in batches, it cannot make good use of the event characteristics of the avatar changing event and cannot more accurately identify abnormal user accounts.

[0004] Therefore, there is a problem in the traditional technology that the accuracy of identifying abnormal user accounts is not high. Summary of the invention

[0005] Based on this, it is necessary to provide an account data processing method, abnormal account identification method, device, computer equipment, computer-readable storage medium and computer program product that can accurately identify account data in an interactive media platform in response to the above-mentioned technical problems.

[0006] A method for processing account data, comprising:

[0007] According to a preset time period, obtaining a changed avatar corresponding to each avatar change event in the interactive media platform;

[0008] Clustering the changed avatars corresponding to each avatar change event to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold;

[0009] Screening out abnormal avatar clusters from each avatar cluster; the number of changed avatars in the abnormal avatar clusters is greater than a preset number threshold;

[0010] The abnormal avatar features of the abnormal avatar clusters are extracted; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

[0011] In one embodiment, clustering the changed avatars corresponding to each avatar change event to obtain avatar clustering results includes:

[0012] Performing vectorization processing on each changed avatar to obtain an avatar vector corresponding to each changed avatar;

[0013] Based on the avatar vectors corresponding to the changed avatars, the changed avatars are clustered to obtain the avatar clustering results.

[0014] In one embodiment, vectorization is performed on each changed avatar to obtain an avatar vector corresponding to each changed avatar, including:

[0015] Scale the changed avatar to obtain a scaled avatar corresponding to the changed avatar;

[0016] The pixel mean of the scaled avatar is obtained, and the avatar vector corresponding to the changed avatar is generated according to the pixel value corresponding to each pixel point in the scaled avatar and the pixel mean.

[0017] In one embodiment, extracting abnormal head portrait features of abnormal head portrait clusters includes:

[0018] The changed avatars in the abnormal avatar cluster are input into the pre-trained image feature extraction model, and the image visual features of each changed avatar in the abnormal avatar cluster are extracted by the pre-trained image visual feature extraction model as the abnormal avatar features of the abnormal avatar cluster.

[0019] In one embodiment, obtaining a changed avatar corresponding to an avatar change event in an interactive media platform according to a preset time period includes:

[0020] In response to each avatar change event of the interactive media platform according to a preset time period, obtain a new avatar download link corresponding to each avatar change event;

[0021] Based on the new avatar download link corresponding to each avatar change event, the changed avatar corresponding to each avatar change event in the interactive media platform is downloaded.

[0022] An abnormal account identification method, comprising:

[0023] Obtain the changed avatar corresponding to the avatar change event in the interactive media platform;

[0024] Extracting the head image features of the changed head image, and matching them with the abnormal head image features to obtain a matching result; the abnormal head image features are determined according to any one of the account data processing methods described above;

[0025] According to the matching result, it is determined whether the user account associated with the avatar change event is an abnormal user account in the interactive media platform.

[0026] An account data processing device, comprising:

[0027] An acquisition module, used to acquire, according to a preset time period, a changed avatar corresponding to each avatar change event in the interactive media platform;

[0028] A clustering module is used to cluster the changed avatars corresponding to each avatar change event to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold;

[0029] A screening module is used to screen out abnormal head portrait clusters from each head portrait cluster; the number of changed head portraits in the abnormal head portrait cluster is greater than a preset number threshold;

[0030] The extraction module is used to extract abnormal avatar features of abnormal avatar clusters; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

[0031] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0032] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0033] A computer program product comprises a computer program, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0034] The above-mentioned account data processing method, device, computer equipment, computer-readable storage medium and computer program product obtain the changed avatars corresponding to each avatar change event in the interactive media platform according to a preset time period; cluster the changed avatars corresponding to each avatar change event to obtain the avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold; abnormal avatar clustering clusters are screened out in each avatar clustering cluster; the number of changed avatars in the abnormal avatar clustering cluster is greater than a preset number threshold; abnormal avatar features of the abnormal avatar clustering cluster are extracted; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform; in this way, it is possible to identify abnormal user accounts in the interactive media platform according to the preset time period. time period, cluster analysis is performed on all avatar change events of the interactive media platform within the current time period to obtain multiple avatar clustering clusters, thereby realizing real-time analysis of avatar change events of the interactive media platform within the current event period, and based on the behavioral characteristics of abnormal users changing avatars in batches in a short period of time, the avatar clustering cluster whose number of changed avatars is greater than a preset number threshold is taken as the abnormal avatar clustering cluster, thereby realizing accurate determination of the avatar of the abnormal user among a large number of changed avatars, and extracting the abnormal avatar features of the abnormal avatar clustering cluster. In the case of subsequent avatar change events, the avatar features of the changed avatar of the avatar change event can be quickly and accurately matched with the abnormal avatar features, thereby improving the efficiency and accuracy of identifying abnormal user accounts. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0036] Figure 1 An application environment diagram of an account data processing method in an embodiment;

[0037] Figure 2 A flowchart of an account data processing method in one embodiment;

[0038] Figure 3 A schematic diagram of a process of an avatar clustering method in an embodiment;

[0039] Figure 4 is a flowchart of an account data processing method in another embodiment;

[0040] Figure 5A flowchart of an abnormal account identification method in one embodiment;

[0041] Figure 6 It is a structural block diagram of an account data processing device in one embodiment;

[0042] Figure 7 It is a structural block diagram of an abnormal account identification device in one embodiment;

[0043] Figure 8 FIG. 4 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0045] The account data processing method provided in the embodiment of the present application can be applied to Figure 1 In the application environment shown. Among them, the terminal 102 communicates with the server 104 through the network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or it can be placed on the cloud or other network servers. Each user account can change the account avatar on the interactive media platform through the terminal 102, and the server 104 obtains the changed avatar corresponding to each avatar change event in the interactive media platform according to the preset time period; the server 104 clusters the changed avatars corresponding to each avatar change event to obtain the avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than the preset similarity threshold; the server 104 selects abnormal avatar clustering clusters from each avatar clustering cluster; the number of changed avatars in the abnormal avatar clustering cluster is greater than the preset number threshold; the server 104 extracts the abnormal avatar features of the abnormal avatar clustering cluster; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart speakers, smart TVs, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. The server 104 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides cloud computing services.

[0046] In an exemplary embodiment, Figure 2 As shown, a method for processing account data is provided, which is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 202 to 208. Among them:

[0047] Step 202: acquiring the changed avatars corresponding to each avatar change event in the interactive media platform according to a preset time period.

[0048] The preset time period may be a preset time interval. In actual applications, the changed avatar corresponding to each avatar change event in the interactive media platform may be obtained every several minutes.

[0049] Among them, interactive media platforms may refer to platforms such as social platforms and game platforms where users can interact.

[0050] Among them, the avatar change event can refer to an event triggered by behaviors such as changing the avatar and creating the avatar. Changing the avatar can be a user replacing an old avatar of an old account, and creating the avatar can be a user uploading an avatar to a newly registered account.

[0051] In actual applications, the present application can also be applied to other scenarios besides the avatar change scenario, such as private chat scenarios or entering a live broadcast room scenario. In the private chat scenario, when the user initiates a private chat, the server will obtain the current avatar of the user account. In the live broadcast room scenario, each time the user enters the live broadcast room, the server will obtain the current avatar of the user account. In different scenarios, after the server obtains the current avatar of the user account, it can execute the method of the present application to identify whether the user account is an abnormal user account by determining whether the current avatar of the user account has abnormal avatar features. An abnormal user account is a user account that "exposes abnormal avatars with abnormal avatar features to other users, thereby achieving abnormal purposes such as traffic diversion."

[0052] The changed avatar may be a new avatar that is replaced after the avatar has changed.

[0053] Optionally, the server obtains the changed avatars corresponding to all avatar change events in the interactive media platform in the current time period every preset time period.

[0054] Step 204, clustering the changed avatars corresponding to each avatar change event to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold.

[0055] The avatar clustering result may include cluster information of multiple avatar clustering clusters. The avatar clustering cluster may include multiple changed avatars. The cluster information of each avatar clustering cluster may include a new avatar download link and a user account identifier corresponding to each changed avatar in the avatar clustering cluster. The new avatar download link may be used to download the changed avatar, and the user account identifier may be used to locate the ID of the user account.

[0056] The preset similarity threshold may be a pre-set value for determining the degree of similarity between different changed avatars. In practical applications, a similarity threshold is first set, and then, in the process of clustering multiple changed avatars, the similarities between the different changed avatars are determined, and the avatars with similarities greater than or equal to the similarity threshold are grouped into one avatar cluster.

[0057] Optionally, the server clusters the changed avatars corresponding to all avatar change events in the interactive media platform in the current time period every preset time period to obtain various avatar clusters corresponding to the current time period.

[0058] Step 206: Filter out abnormal avatar clusters from each avatar cluster; the number of changed avatars in the abnormal avatar cluster is greater than a preset number threshold.

[0059] Among them, the abnormal avatar clustering cluster can be an avatar clustering cluster in which the number of changed avatars is greater than a preset number threshold, each changed avatar in the abnormal avatar clustering cluster is regarded as a changed avatar of an abnormal user account, and the cluster information of the abnormal avatar clustering cluster can include the user account identifiers corresponding to each changed avatar in the abnormal avatar clustering cluster. According to the user account identifiers corresponding to each changed avatar in the abnormal avatar clustering cluster, the abnormal user accounts in the interactive media platform can be located.

[0060] The preset number threshold may be a pre-set value for determining whether an avatar cluster is an abnormal avatar cluster. When the number of avatars in an avatar cluster after change is greater than the preset number threshold, the avatar cluster may be determined as an abnormal avatar cluster.

[0061] Optionally, the server determines the number of changed avatars in each avatar clustering cluster, and determines the avatar clustering cluster in which the number of changed avatars is greater than a preset number threshold as an abnormal avatar clustering cluster.

[0062] Step 208, extracting abnormal avatar features of the abnormal avatar clusters; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

[0063] Among them, the abnormal avatar features may be image features of the avatar used by the abnormal user account, for example, visual features.

[0064] Among them, an abnormal user account may be a user account that has the behavioral characteristic of "changing profile pictures in batches within a short period of time" on an interactive media platform.

[0065] Optionally, the server regards all changed avatars in the abnormal avatar cluster as avatars of abnormal user accounts, and extracts features of all avatars in the abnormal avatar cluster as abnormal avatar features.

[0066] In the above account data processing method, the changed avatars corresponding to each avatar change event in the interactive media platform are obtained according to a preset time period; the changed avatars corresponding to each avatar change event are clustered to obtain the avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold; abnormal avatar clustering clusters are screened out from each avatar clustering cluster; the number of changed avatars in the abnormal avatar clustering cluster is greater than a preset number threshold; abnormal avatar features of the abnormal avatar clustering cluster are extracted; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform; in this way, the interactive media platform can be screened according to the preset time period. All avatar change events within the current time period are clustered and analyzed to obtain multiple avatar clustering clusters, thereby realizing real-time analysis of avatar change events within the current event period of the interactive media platform. Based on the behavioral characteristics of abnormal users changing avatars in batches in a short period of time, the avatar clustering clusters in which the number of changed avatars is greater than a preset number threshold can be used as abnormal avatar clustering clusters, thereby realizing accurate determination of the avatars of abnormal users among a large number of changed avatars, and extracting abnormal avatar features of abnormal avatar clustering clusters. In the event of a subsequent avatar change event, the avatar features of the changed avatar of the avatar change event can be quickly and accurately matched with the abnormal avatar features, thereby improving the efficiency and accuracy of identifying abnormal user accounts.

[0067] In an exemplary embodiment, Figure 3 As shown, step 204 includes steps 302 to 304 .

[0068] in:

[0069] Step 302, vectorization processing is performed on each changed avatar to obtain an avatar vector corresponding to each changed avatar.

[0070] The avatar vector corresponding to the changed avatar may be a hash vector obtained by hash mapping the changed avatar, and the avatar vector has the position information of each pixel of the changed avatar. In practical applications, the image features of the changed avatar extracted by the deep learning model may also be used as the avatar vector. The avatar vector obtained by the deep learning model can more accurately represent the avatar features of the changed avatar, but it requires more computing resources in practical applications.

[0071] Optionally, the server performs vectorization processing on the changed avatars corresponding to each avatar change event to obtain avatar vectors corresponding to each changed avatar.

[0072] Step 304: clustering the changed avatars based on the avatar vectors corresponding to the changed avatars to obtain avatar clustering results.

[0073] Optionally, the server performs clustering processing on each changed avatar based on the avatar vector corresponding to each changed avatar to obtain at least one avatar cluster.

[0074] In this embodiment, each changed avatar is vectorized to obtain the avatar vector corresponding to each changed avatar, and then based on the avatar vector corresponding to each changed avatar, each changed avatar is clustered to obtain the avatar clustering result. The similarity between the avatar vectors can be used to accurately cluster the changed avatars, thereby improving the accuracy of avatar clustering.

[0075] In an exemplary embodiment, each changed avatar is vectorized to obtain an avatar vector corresponding to each changed avatar, including: scaling the changed avatar to obtain a scaled avatar corresponding to the changed avatar; obtaining a pixel mean of the scaled avatar, and generating an avatar vector corresponding to the changed avatar based on the pixel value corresponding to each pixel point in the scaled avatar and the pixel mean.

[0076] The scaled avatar may refer to an avatar obtained by scaling the changed avatar.

[0077] The pixel mean may refer to a value obtained by averaging the pixel values ​​of each pixel point.

[0078] Optionally, the server scales the changed avatar to obtain a scaled avatar corresponding to the changed avatar, and then obtains a pixel mean of the scaled avatar, and generates a hash vector corresponding to the changed avatar based on the pixel value corresponding to each pixel point in the scaled avatar and the pixel mean.

[0079] In practical applications, since the sizes of different modified avatars may vary, to unify the input of images, the server uniformly scales the sizes of all modified avatars to 8*8. Each modified avatar corresponds to 64 pixel points. Since some of the modified images are single-channel grayscale images, some are RGB three-channel color images, and some are RGBA four-channel color images, to unify the next input standard, the server converts non-single-channel images into single-channel grayscale images, thereby converting all modified avatars into grayscale images. Each modified avatar can obtain an 8x8 integer matrix G. For an integer matrix G of a modified avatar, the server calculates the average value of all elements in matrix G, that is, the pixel average value a. According to the pixel average value a, the fingerprint of the modified avatar is calculated. The ahash of the input image is initialized as "", and each pixel point of matrix G is traversed row by row from left to right. If the element G(i, j) in the i-th row and j-th column >= a, then ahash += "1"; if the element G(i, j) in the i-th row and j-th column < a, then ahash += "0", until the ahash value of the modified avatar is finally obtained. Subsequently, the Hamming distance between the ahash values of two modified avatars can be determined, and two images with a Hamming distance less than 10 are determined as similar images.

[0080] In this embodiment, by scaling the modified avatar, a scaled avatar corresponding to the modified avatar is obtained, and then the pixel average value of the scaled avatar is obtained. According to the pixel values corresponding to each pixel point in the scaled avatar and the pixel average value, an avatar vector corresponding to the modified avatar is generated, which can quickly obtain the avatar vector of the modified avatar, thereby facilitating the subsequent rapid clustering of each modified avatar.

[0081] In an exemplary embodiment, the abnormal avatar features of the abnormal avatar clustering cluster are extracted, including: inputting each modified avatar in the abnormal avatar clustering cluster into a pre-trained image feature extraction model, and extracting the image visual features of each modified avatar in the abnormal avatar clustering cluster through the pre-trained image visual feature extraction model as the abnormal avatar features of the abnormal avatar clustering cluster.

[0082] Among them, the pre-trained image feature extraction model can be a pre-trained deep learning model, which can be used to extract the image features in the avatar. The avatar vector in the above embodiment represents the position information of each pixel point of the avatar, and the image features here represent visual features, including visual elements such as color, texture, and shape. These are the basic composition information of the picture, and at the same time include the global information of the avatar (such as the overall layout and scene type) and local information (such as the details of a specific area).

[0083] Optionally, the server inputs each changed avatar in the abnormal avatar clustering cluster into a pre-trained image feature extraction model, and then extracts the image visual features of each changed avatar in the abnormal avatar clustering cluster through the pre-trained image visual feature extraction model as the abnormal avatar features of the abnormal avatar clustering cluster.

[0084] In this embodiment, by inputting each changed avatar in the abnormal avatar clustering cluster into a pre-trained image feature extraction model, and extracting the image visual features of each changed avatar in the abnormal avatar clustering cluster through the pre-trained image visual feature extraction model, as the abnormal avatar features of the abnormal avatar clustering cluster, the pre-trained image visual feature extraction model can be used to accurately extract the abnormal avatar features of the abnormal avatar clustering cluster. Subsequently, when any changed avatar is obtained, the pre-trained image visual feature extraction model can also be used to extract the avatar features of any changed avatar, and then the abnormal avatar features extracted by the image visual feature extraction model and the avatar features of any changed avatar are accurately compared, which is conducive to accurately identifying abnormal avatars.

[0085] In an exemplary embodiment, a changed avatar corresponding to an avatar change event in an interactive media platform is obtained according to a preset time period, including: according to the preset time period, in response to each avatar change event of the interactive media platform, obtaining a new avatar download link corresponding to each avatar change event; based on the new avatar download link corresponding to each avatar change event, downloading the changed avatar corresponding to each avatar change event in the interactive media platform.

[0086] The new avatar download link may refer to a download link corresponding to the changed avatar.

[0087] Optionally, the server responds to each avatar change event of the interactive media platform according to a preset time period and records the event information of each avatar change event. The event message may be a new avatar download link corresponding to the avatar change event. The server downloads the changed avatar corresponding to each avatar change event based on the new avatar download link corresponding to each avatar change event.

[0088] In this embodiment, by responding to each avatar change event of the interactive media platform according to a preset time period, a new avatar download link corresponding to each avatar change event is obtained, and then based on the new avatar download link corresponding to each avatar change event, the changed avatar corresponding to each avatar change event in the interactive media platform is downloaded. The changed avatar corresponding to each avatar change event in the interactive media platform can be downloaded according to the preset time period, thereby realizing real-time downloading of the changed avatar.

[0089] In an exemplary embodiment, Figure 4 As shown, a method for processing account data is provided, which is applied to Figure 1Taking the server 104 in the example as an example, the following steps are included:

[0090] Step 402 : In response to each avatar change event of the interactive media platform according to a preset time period, a new avatar download link corresponding to each avatar change event is obtained.

[0091] Step 404 , based on the new avatar download link corresponding to each avatar change event, download the changed avatar corresponding to each avatar change event in the interactive media platform.

[0092] Step 406, clustering the changed avatars corresponding to each avatar change event to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold.

[0093] Step 408: Filter out abnormal avatar clusters from each avatar cluster; the number of changed avatars in the abnormal avatar cluster is greater than a preset number threshold.

[0094] Step 410, extracting abnormal avatar features of abnormal avatar clusters; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

[0095] It should be noted that the specific limitations of the above steps can refer to the specific limitations of an account data processing method above.

[0096] In an exemplary embodiment, Figure 5 As shown, a method for identifying abnormal accounts is provided, and the method is applied to Figure 1 The server 104 in the example is used as an example to illustrate, including the following steps 502 to 506. Among them:

[0097] Step 502: Obtain a changed avatar corresponding to the avatar change event in the interactive media platform.

[0098] Optionally, the user account changes the avatar in the interactive media platform, and the server responds to the avatar change operation of the user account and obtains event information of an avatar change event triggered by the avatar change operation, the event information including a new avatar download link corresponding to the avatar change event, and the server downloads the changed avatar corresponding to the avatar change event based on the new avatar download link.

[0099] Step 504, extracting the avatar features of the changed avatar and matching them with the abnormal avatar features to obtain a matching result; the abnormal avatar features are determined according to any of the account data processing methods described above.

[0100] Optionally, the server extracts avatar features of the changed avatar through a pre-trained image feature extraction model, and matches the avatar features of the changed avatar with various abnormal avatar features pre-stored in a feature database to obtain a matching result.

[0101] Step 506: Determine, based on the matching result, whether the user account associated with the avatar change event is an abnormal user account in the interactive media platform.

[0102] Optionally, when the matching result indicates that the avatar features of the avatar after the change are similar to a certain type of abnormal avatar features, the server determines the user account that triggers the avatar change event as an abnormal user account.

[0103] In the above abnormal account identification method, the changed avatar corresponding to the avatar change event in the interactive media platform is obtained, the avatar features of the changed avatar are extracted, and matched with the abnormal avatar features to obtain a matching result. According to the matching result, it is determined whether the user account associated with the avatar change event is an abnormal user account in the interactive media platform. By quickly matching the avatar features, a large number of avatar change events can be quickly processed, thereby improving the accuracy of identifying abnormal user accounts.

[0104] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0105] Based on the same inventive concept, the embodiment of the present application also provides an account data processing device for implementing the account data processing method involved above and an abnormal account identification device for implementing the abnormal account identification method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more account data processing device embodiments provided below can refer to the above limitations on the account data processing method, and the specific limitations in one or more abnormal account identification device embodiments provided below can refer to the above limitations on the abnormal account identification method, which will not be repeated here.

[0106] In an exemplary embodiment, Figure 6As shown, an account data processing device is provided, including: an acquisition module 602, a clustering module 604, a screening module 606 and an extraction module 608, wherein:

[0107] An acquisition module 602 is used to acquire, according to a preset time period, a changed avatar corresponding to each avatar change event in the interactive media platform;

[0108] The clustering module 604 is used to cluster the changed avatars corresponding to each avatar change event to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold;

[0109] A screening module 606 is used to screen out abnormal avatar clusters from each avatar cluster; the number of changed avatars in the abnormal avatar cluster is greater than a preset number threshold;

[0110] The extraction module 608 is used to extract abnormal avatar features of abnormal avatar clusters; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

[0111] In one embodiment, the clustering module 604 is specifically used to perform vectorization processing on each changed avatar to obtain an avatar vector corresponding to each changed avatar; based on the avatar vector corresponding to each changed avatar, cluster each changed avatar to obtain an avatar clustering result.

[0112] In one embodiment, the clustering module 604 is specifically used to scale the changed avatar to obtain a scaled avatar corresponding to the changed avatar; obtain the pixel mean of the scaled avatar, and generate an avatar vector corresponding to the changed avatar based on the pixel value corresponding to each pixel point in the scaled avatar and the pixel mean.

[0113] In one of the embodiments, the extraction module 608 is specifically used to input each changed avatar in the abnormal avatar clustering cluster into a pre-trained image feature extraction model, and extract the image visual features of each changed avatar in the abnormal avatar clustering cluster through the pre-trained image visual feature extraction model as the abnormal avatar features of the abnormal avatar clustering cluster.

[0114] In one of the embodiments, the acquisition module 602 is specifically used to respond to each avatar change event of the interactive media platform according to a preset time period, obtain a new avatar download link corresponding to each avatar change event; based on the new avatar download link corresponding to each avatar change event, download the changed avatar corresponding to each avatar change event in the interactive media platform.

[0115] In an exemplary embodiment, Figure 7As shown, an abnormal account identification device is provided, including: an avatar acquisition module 702, a feature matching module 704 and an account determination module 706, wherein:

[0116] An avatar acquisition module 702 is used to acquire a changed avatar corresponding to an avatar change event in an interactive media platform;

[0117] The feature matching module 704 is used to extract the avatar features of the changed avatar and match them with the abnormal avatar features to obtain a matching result; the abnormal avatar features are determined according to any of the account data processing methods described above;

[0118] The account determination module 706 is used to determine whether the user account associated with the avatar change event is an abnormal user account in the interactive media platform according to the matching result.

[0119] Each module in the above-mentioned account data processing device and abnormal account identification device can be implemented in whole or in part by software, hardware and their combination. Each module can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module above.

[0120] In an exemplary embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 8 As shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, referred to as I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store account data processing data and abnormal account identification data. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, an account data processing method and an abnormal account identification method are implemented.

[0121] Those skilled in the art will understand that Figure 8The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0122] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the above-mentioned account data processing method and abnormal account identification method. Here, the steps of the account data processing method and abnormal account identification method can be the steps of the account data processing method and abnormal account identification method in each of the above-mentioned embodiments.

[0123] In one embodiment, a computer-readable storage medium is provided, which stores a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned account data processing method and abnormal account identification method. Here, the steps of the account data processing method and abnormal account identification method can be the steps of the account data processing method and abnormal account identification method in each of the above-mentioned embodiments.

[0124] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by a processor, the processor executes the steps of the above-mentioned account data processing method and abnormal account identification method. Here, the steps of the account data processing method and abnormal account identification method can be the steps of the account data processing method and abnormal account identification method in each of the above-mentioned embodiments.

[0125] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0126] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0127] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be construed as limiting the scope of the present application. It should be noted that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for processing account data, characterized in that: The method comprises: According to a preset time period, obtaining a changed avatar corresponding to each avatar change event in the interactive media platform; Clustering the changed avatars corresponding to each of the avatar change events to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold; Screening out abnormal avatar clusters from each of the avatar clusters; the number of the changed avatars in the abnormal avatar clusters is greater than a preset number threshold; The abnormal avatar features of the abnormal avatar cluster are extracted; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

2. The method according to claim 1, characterized in that The step of clustering the changed avatars corresponding to the avatar change events to obtain avatar clustering results includes: Performing vectorization processing on each of the changed avatars to obtain an avatar vector corresponding to each of the changed avatars; Based on the avatar vectors corresponding to the changed avatars, the changed avatars are clustered to obtain the avatar clustering result.

3. The method according to claim 2, characterized in that The vectorization processing is performed on each of the changed avatars to obtain the avatar vector corresponding to each of the changed avatars, including: Scaling the changed avatar to obtain a scaled avatar corresponding to the changed avatar; The pixel mean of the scaled avatar is obtained, and an avatar vector corresponding to the changed avatar is generated according to the pixel value corresponding to each pixel point in the scaled avatar and the pixel mean.

4. The method according to claim 1, characterized in that The step of extracting the abnormal head portrait features of the abnormal head portrait cluster includes: The changed avatars in the abnormal avatar cluster are input into a pre-trained image feature extraction model, and the image visual features of each changed avatar in the abnormal avatar cluster are extracted by the pre-trained image visual feature extraction model as the abnormal avatar features of the abnormal avatar cluster.

5. The method according to claim 1, characterized in that The step of obtaining the changed avatar corresponding to the avatar change event in the interactive media platform according to a preset time period includes: In response to each of the avatar change events of the interactive media platform according to a preset time period, obtaining a new avatar download link corresponding to each of the avatar change events; Based on the new avatar download link corresponding to each avatar change event, the changed avatar corresponding to each avatar change event in the interactive media platform is downloaded.

6. A method for identifying abnormal accounts, characterized in that: The method comprises: Obtain the changed avatar corresponding to the avatar change event in the interactive media platform; Extracting the head image features of the changed head image and matching them with the abnormal head image features to obtain a matching result; the abnormal head image features are determined according to the account data processing method according to any one of claims 1 to 5; According to the matching result, it is determined whether the user account associated with the avatar change event is an abnormal user account in the interactive media platform.

7. An account data processing device, characterized in that: The device comprises: An acquisition module, used to acquire, according to a preset time period, a changed avatar corresponding to each avatar change event in the interactive media platform; A clustering module, used for clustering the changed avatars corresponding to each of the avatar change events to obtain an avatar clustering result; the avatar clustering result includes cluster information of at least one avatar clustering cluster; the similarity between different changed avatars belonging to the same avatar clustering cluster is greater than a preset similarity threshold; A screening module, used for screening out abnormal head portrait clusters from each head portrait cluster; the number of the changed head portraits in the abnormal head portrait cluster is greater than a preset number threshold; The extraction module is used to extract the abnormal avatar features of the abnormal avatar cluster; the abnormal avatar features are used to identify abnormal user accounts in the interactive media platform.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images