WO2019165697A1

WO2019165697A1 - Method and device for identifying click farming users, terminal device and storage medium

Info

Publication number: WO2019165697A1
Application number: PCT/CN2018/084638
Authority: WO
Inventors: 王璐; 陈少杰; 张文明
Original assignee: 武汉斗鱼网络科技有限公司
Priority date: 2018-02-28
Filing date: 2018-04-26
Publication date: 2019-09-06
Also published as: CN108390883B; CN108390883A

Abstract

The present disclosure relates to the technical field of the Internet, in particular to a method and device for identifying click farming users, a terminal device and a storage medium, the method for identifying a click farming user comprising: acquiring user features of all users; determining a first risk score according to the user features corresponding to each user; further identifying a click farming user determined among all users as a target click farming user; identifying a non-click farming user determined among all users as a target normal user; determining a second risk score of each user according to the target click farming user and the target normal user; and finally, determining the final risk score of each user being a click farming user according to the first risk score and the second risk score, and identifying click farming users according to the final risk score. With the present solution, the accuracy of identifying a user as a click farming user is improved.

Description

Method, device, terminal device and storage medium for identifying popular users

Cross-reference to related applications

This application claims priority to Chinese Patent Application No. 2018101691861, entitled "Identification Method, Device and Terminal Device for Popularity Users", filed on February 28, 2018, the entire contents of which are incorporated herein by reference. In the application.

Technical field

The present disclosure relates to the field of Internet technologies, and in particular, to a method, an apparatus, a terminal device, and a storage medium for identifying a popular user.

Background technique

On many Internet platforms, in order to achieve certain purposes, there are often false popular behaviors, such as Taobao brushing or live broadcast platform to the main broadcast, etc., such popular behavior has a great impact on the ecological environment of the Internet platform.

The current method of identifying abnormal popularity of brushing behavior is to adopt some strong rules. These strong rules are identified by some obvious abnormal features. This method can identify some risk users, but the requirements of this method are very strict, so only A cheat user with obvious characteristics can be found, and users who are not obvious in characteristics but have cheating behavior will be omitted. Therefore, it is necessary to provide a method for identifying popular users with high accuracy.

Summary of the invention

The purpose of the present disclosure includes a method of identifying a popular user to improve the accuracy of the user who identifies the brush.

Another object of the present disclosure includes an identification device for a popular user to improve the accuracy of identifying a popular user of the brush.

Another object of the present disclosure includes a terminal device to improve the accuracy of identifying a popular user of a brush.

In order to achieve the above object, the technical solution adopted by the embodiment of the present disclosure is as follows:

In a first aspect, an embodiment of the present disclosure provides a method for identifying a popular user, the method comprising: acquiring user characteristics of all users; determining a first risk score according to user characteristics corresponding to each user; and identifying all users The popular user is a popular user for the target; the non-brushing popular user determined among all the users is the target normal user; and the second risk score of each user is determined according to the user characteristics corresponding to the target popular user and the target normal user; Determining, according to the first risk score and the second risk score, a final risk score of each user as a popular user; and identifying other popular users according to the final risk score.

In a second aspect, an embodiment of the present disclosure further provides an identification device for a popular user, the device comprising: an acquisition module configured to acquire user characteristics of all users; and a first determining module configured to correspond to each user The user feature determines a first risk score; the first identification module is configured to identify that the brushed popularity user among all the users is the target popularity user; and the second identification module is configured to identify the non-brush popularity user determined by all the users as the target normal a second determining module, configured to determine a second risk score of each user according to the user characteristics corresponding to the target brush popularity user and the target normal user; the score determining module configured to perform according to the first risk score and the The second risk score determines each user's final risk score for the popular user; and identifies other popular users based on the final risk score.

In a third aspect, an embodiment of the present disclosure further provides a terminal device, the terminal device including a memory and a processor, the memory configured to store computer program code, the processor configured to execute stored in the memory The computer program code implements the method of identifying the popular user.

In a fourth aspect, an embodiment of the present disclosure further provides a terminal device, a readable storage medium, storing executable instructions, where the instructions are implemented when executed by one or more processors User identification method.

An embodiment of the present disclosure provides a method, an apparatus, a terminal device, and a storage medium for identifying a popular user, and the method and device for identifying the popular user are applied to the terminal device. The method for identifying popular users includes obtaining user characteristics of all users, determining a first risk score according to user characteristics corresponding to each user, and then identifying a popular user determined by all users as a target popular user, and identifying all users. The non-brushing popular user is the target normal user, and the second risk score of each user is determined according to the target brushing popularity user and the target normal user. Finally, the final risk score of each user is determined according to the first risk score and the second risk score. In this solution, the first risk score and the second risk score of the current user are respectively calculated by two aspects, and then the final risk score of the current user is calculated according to the first risk score and the second risk score, which may In the first risk score and the second risk score, the popular popularity of the user is determined, which avoids the omission of the popular user, and improves the accuracy of identifying the user as a popular user.

The above described objects, features, and advantages of the present invention will become more apparent from the description of the appended claims.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings to be used in the embodiments will be briefly described below. It should be understood that the following drawings show only certain embodiments of the present disclosure, and thus It should be seen as a limitation on the scope, and those skilled in the art can obtain other related drawings according to these drawings without any creative work.

FIG. 1 is a schematic structural diagram of a terminal device according to an embodiment of the present disclosure.

FIG. 2 is a schematic flow chart of a method for identifying a popular user by providing an embodiment of the present disclosure.

FIG. 3 is a schematic flow chart showing a sub-step of a method for identifying a popular user by providing an embodiment of the present disclosure.

FIG. 4 is a schematic flow chart showing another sub-step of a method for identifying a popular user by providing an embodiment of the present disclosure.

FIG. 5 is a schematic diagram of functional modules of an identification device for a popular user of the present disclosure.

Illustration: 100-terminal device; 110-identification device for brushing popular users; 120-memory; 130-storage controller; 140-processor; 150-peripheral interface; 160-input-output unit; 170-audio unit; a display unit; 111-acquisition module; 112-first determination module; 113-first identification module; 114-second identification module; 115-second determination module; 116-score determination module.

Detailed ways

The technical solutions in the embodiments of the present disclosure are clearly and completely described in the following with reference to the accompanying drawings in the embodiments of the present disclosure. It is obvious that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. The components of the disclosed embodiments, which are generally described and illustrated in the figures herein, can be arranged and designed in various different configurations. The detailed description of the embodiments of the present disclosure, which is set forth in the claims All other embodiments obtained by a person skilled in the art based on the embodiments of the present disclosure without creative efforts are within the scope of the present disclosure.

It should be noted that similar reference numerals and letters indicate similar items in the following figures, and therefore, once an item is defined in a drawing, it is not necessary to further define and explain it in the subsequent drawings. Meanwhile, in the description of the present disclosure, the terms "first", "second", and the like are only configured to distinguish descriptions, and are not to be construed as indicating or implying relative importance.

FIG. 1 is a schematic structural diagram of a terminal device 100 according to an embodiment of the present disclosure. A method for identifying a popular user is applied to the terminal device 100, and the terminal device 100 may be, but not Limited to smart electronic devices such as tablets and desktop computers. The terminal device 100 includes an identification device 110 for brushing popular users, a memory 120, a memory controller 130, a processor 140, a peripheral interface 150, an input and output unit 160, an audio unit 170, and a display unit 180.

The components of the memory 120, the memory controller 130, the processor 140, the peripheral interface 150, the input and output unit 160, the audio unit 170, and the display unit 180 are directly or indirectly electrically connected to each other to implement data transmission or Interaction. For example, the components can be electrically connected to one another via one or more communication buses or signal lines. The identification device 110 of the popular user includes at least one software function module that can be stored in the memory 120 or is solidified in an operating system (OS) of the terminal device 100 in the form of software or firmware. . The processor 140 is configured to execute an executable module stored in the memory 120, such as a software function module or a computer program included in the identification device 110 of the popular user.

The memory 120 can be, but not limited to, a random access memory (RAM), a read only memory (ROM), and a programmable read-only memory (PROM). Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), and the like. The memory 120 is configured to store a program, and the processor 140 executes the program after receiving the execution instruction, and the method executed by the terminal device 100 defined by the flow process disclosed in any of the foregoing embodiments of the present disclosure may be applied. In processor 140, or implemented by processor 140.

Processor 140 may be an integrated circuit chip with signal processing capabilities. The processor 140 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP Processor, etc.), or a digital signal processor (DSP), an application specific integrated circuit. (ASIC), Field Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware component. The methods, steps, and logical block diagrams disclosed in the embodiments of the present disclosure may be implemented or carried out. The general purpose processor may be a microprocessor or the processor 140 may be any conventional processor 140 or the like.

The peripheral interface 150 couples various input/output devices to the processor 140 and to the memory 120. In some embodiments, peripheral interface 150, processor 140, and memory controller 130 can be implemented in a single chip. In other instances, they can be implemented by separate chips.

The input output unit 160 is configured to provide input to the user to enable user interaction with the terminal device 100. The input and output unit 160 may be, but not limited to, a mouse, a keyboard, and the like.

The audio unit 170 provides an audio interface to the user, which may include one or more microphones, one or more speakers, and audio circuitry.

The display unit 180 provides an interactive interface (eg, a user interface) between the terminal device 100 and the user or is configured to display image data to the user for reference. In this embodiment, the display unit 180 can be a liquid crystal display or a touch display. For a touch display, it can be a capacitive touch screen or a resistive touch screen that supports single-point and multi-touch operations. Supporting single-point and multi-touch operations means that the touch display can sense simultaneous touch operations from one or more locations on the touch display, and the touch operation is transferred to the processor 140. Perform calculations and processing.

2 is a schematic flowchart of a method for identifying a popular user of the present disclosure. The method for identifying a popular user may be configured to identify a user who has a billing behavior in each major website. The website can be, but is not limited to, an e-commerce website (such as Taobao), a live broadcast platform, and the like. In the embodiment of the present disclosure, the identification method of the popular user is applied to the live broadcast platform as an example for description. The method includes:

Step S110, acquiring user characteristics of all users.

It should be noted that the user feature is a user feature related to the user's popular behavior. For example, in the live broadcast platform, the user usually broadcasts popularity in a plurality of live broadcast rooms or in a short time in a plurality of live broadcast rooms, and the number of live broadcast rooms or users that the user watches during the predetermined time is The number of bullets that are played in multiple live broadcasts within a predetermined time period is the user feature that needs to be acquired. It is easy to understand that each user has multiple user features. In order to facilitate the clear description of the solution provided by the embodiment, so that those skilled in the art can clearly understand, the user feature can be selected for each time when the user views multiple live rooms in a short time. The data, such as the start time and end time of each live room watch, and the watch time for each live room, for example, the user feature of user A may be: user A watching the start time M1 of the live room X, the end time M2 and the viewing time M3.

Further, the terminal device may acquire user characteristics of all users, for example, three ways may be adopted.

As a first implementation manner, the terminal device may obtain a user log generated during a live broadcast of a live broadcast, where the user log indicates that each user who views the live broadcast room during the entire live broadcast process is in the entire viewing process. All the viewings of the live broadcast room, including the live broadcast room, can obtain the start time and end time of each user watching each live broadcast room, as well as the viewing time, and then obtain all the viewing of the live broadcast room. User characteristics for all users.

As a second implementation manner, the terminal device may obtain the duration of the time according to a preset, and intermittently obtain a user log generated during a preset duration of the live broadcast during a live broadcast, where The user log indicates that each user who views the live broadcast room during the preset acquisition time has watched all the viewed live broadcast rooms including the live broadcast room during the entire viewing process, that is, every preset time period can be obtained. Each user views the start time and end time of each live broadcast room, as well as the viewing time, and then obtains the user characteristics of all users who view the live broadcast room. For example, if the preset acquisition time is 2 hours and the preset time interval is 30 minutes, the terminal device continuously obtains the user log generated within the 2 hours during a certain 2 hours during the live broadcast of the live broadcast. Intermittently for a period of 30 minutes, and after 30 minutes of interruption, continue to obtain the user log generated by the next 2 hours, thereby forming a loop and ending the broadcast. It can be understood that the method can effectively reduce the processing capacity of the terminal device for data, and can reduce the power consumption to a certain extent.

As a third implementation manner, the terminal device can first monitor the current popularity value between the live broadcasts, so that the current popularity value of the live broadcast room can be obtained in real time. The terminal device can determine whether the current popularity value is less than the popularity threshold by passing a preset popularity threshold. When it is judged to be not less than, it means that the popularity of the live broadcast room is relatively high, and the number of viewing users in the live broadcast room is also sufficient, and the attention of the live broadcast room is relatively high (for example, a big anchor), and then the live broadcast room Popularity is likely to be based on the influence of the anchor, while popularity may be a small probability event. Therefore, the terminal device can proceed without further processing and continue monitoring. When it is judged to be less than, it means that the popularity of the live broadcast room is relatively low, and the number of viewing users in the live broadcast room is relatively small, and the attention degree of the live broadcast room is relatively low (for example, a small anchor), and then the live broadcast room has Perhaps Rong Ge brushed popularity to provide attention. Therefore, the terminal device can obtain the user log generated by the live broadcast room from this time, and until the live broadcast of the live broadcast room ends, wherein the user log indicates that each live broadcast room is viewed during the process from the start of monitoring to the end of the live broadcast. During the entire viewing process, the user can view the viewing time of all the live broadcast rooms including the live broadcast room, that is, the start time and end time of each user watching each live broadcast room, and the viewing time, and then the viewing time. User characteristics of all users viewing the live room were also obtained. It can be understood that, in this manner, the terminal device performs the process when it is required to acquire the user features of all users, thereby effectively reducing the processing capacity of the terminal device for data, and also reducing power consumption to a certain extent. effect.

It should be noted that, in the third embodiment, according to the actual situation, for example, when the live broadcast room with high popularity is likely to be popular, the terminal device may also be adjusted to monitor when the current popularity value is greater than the popularity threshold. . That is to say, the embodiment provided in this embodiment is only an optional manner, and is not limited thereto, and the specific implementation may be selected according to actual conditions. Step S120: Determine a first risk score according to a user characteristic corresponding to each user.

That is to say, the first risk score of each user is judged according to the acquired user characteristics, and the possibility of the user being the popular user is determined according to the first risk score. Please refer to FIG. 3 , which is a schematic flowchart of a sub-step of step S120 of the method for identifying a popular user of the present disclosure. The step S120 includes:

Step S121, classifying all users according to the user feature by using a predetermined algorithm.

The predetermined algorithm is a k-means algorithm, which is a clustering algorithm, and the effect of grouping is realized by continuous iteration of the center point and distance calculation. Further, according to the acquired user characteristics, the user is divided into 10 categories according to the k-means algorithm, which is easy to understand, and the specific division can be customized according to the needs of the user.

Step S122, selecting key indicators in the user characteristics corresponding to each type of user.

That is to say, users classified into one class have similar user characteristics, and then select key indicators for each user feature corresponding to each type of user, for example, the key indicator may be the user's viewing time for each live broadcast, etc. Preferably, the key indicator is multiple, that is, the key indicator is a plurality of viewing durations corresponding to multiple live broadcasts during the entire process of viewing multiple live broadcasts. In step S123, an average value of key indicators of users under each category is calculated.

Step S124, determining a first risk score of each user according to the size of the average value.

That is to say, if the selected key indicator is the user's characteristics such as the viewing time of each live broadcast room, the shorter the duration, the higher the probability that the user is popular for the main broadcast, and then the average of the multiple key indicators. The smaller the value, the higher the risk that the user is a popular user, and the high risk category is recorded as 1 point, and the low risk category is recorded as 0.1 point. In turn, all categories of users will determine the first risk score based on the category they belong to.

In step S130, it is identified that the popular user identified by all the users is the target popular user.

That is to say, the user who selects the user feature to meet the preset strong rule is the target user, and the preset strong rule requires the user to meet the predetermined strict restriction feature, such as, but not limited to, a certain feature. The user has changed multiple devices to log in to his account within a predetermined time, or for the live broadcast platform, there is no log for the user in the background, but the user can continue to play the bullet. The preset strong rule indicates that the user has an obvious popularity feature, and if a user has a user feature that meets a preset strong rule, it is determined that the user is a target popular user. It can be determined that the user who meets the preset strong rule must be a popular user for the target, but it cannot be excluded that the user who does not meet the preset strong rule must not be the target user, because the preset strong rule can only be filtered out. Users with obvious popularity, but can not rule out the suspicion of other users.

Step S140, identifying that the non-brushing popular users determined among all the users are the target normal users.

That is to say, a normal user among all users can be identified according to a predetermined rule, the predetermined rule includes a plurality of user features, and the plurality of user features jointly determine that the corresponding user is a target normal user, wherein determining that one user is normal The most important user characteristic of the user is the payment behavior, that is, the behavior of determining whether the user has continuously paid for the service enjoyed. The normal user refers to a user who determines that there is no popular behavior.

Step S150: Determine a second risk score of each user according to the user characteristics corresponding to the target brush popularity user and the target normal user.

Specifically, please refer to FIG. 4 , which is a schematic flowchart of a sub-step of step S150 of the method for identifying a popular user of the present disclosure. The step S150 includes:

In step S151, a network structure between all users is constructed.

It is easy to understand that each user has a unique IP or a fixed device address when logging in to the website or live broadcast platform, but there are cases where multiple users have the same IP address or device address due to the presence of a popular user. Thus, users with the same IP address or device address are connected by a straight line, thereby forming a network structure among all users.

Step S152: Calculate the similarity weight between each two users according to the user characteristics of the target brush popularity user and the target normal user.

Since the target brush popularity user and the target normal user have been determined in advance, the user characteristics describing the user as the target brush popularity user have been determined, and the user characteristics describing the user as the target normal user have also been determined. If, in the network structure, a user is connected to the target user and/or the target normal user, the same user feature of the user and the target user is selected, and the similarity right between the user and the target user is calculated. The value, the same user feature of the user and the target normal user is selected, and the similarity weight of the user and the target popular user is calculated.

If it is not determined whether there is a suspected user A and user B, user C and user D are connected in the network structure, wherein both user B and user C are determined target popularity users, and user D is the determined target normal user. , respectively, the similarity between user A and users B, C, and D is calculated, and the calculation result is W(AB)=0.7W(AC)=0.5W(AD)=0.3, that is, user A and user The similarity between B is 0.7, the similarity between user A and user C is 0.5, and the similarity between user A and user D is 0.3. The similarity is calculated as:

Among them, Xui is the i-th user feature of user u, Xvi is the i-th user feature of user v, and w _uv is the similarity weight between user u and user v.

It should be noted that since only the target brush popularity user and the target normal user are determined, at the beginning of the calculation, even if there is the unknown or not, the brush popularity user has the same as other unknown users, the target brush popularity user, and the target normal user. The connection relationship only calculates whether the unknown has a similarity between the popular user and the target brush user and the target normal user. Then, by using the algorithm to make the similarity weight of the currently unknown user known, the similarity weights of other users can be calculated according to the connection relationship between the currently unknown user and other users in the network structure.

Step S153: Calculate a second risk score of the current user according to a similarity weight of other users connected to the current user in the network structure, and an initial score of the target brush popularity user and the target normal user.

That is to say, in the case of knowing the similarity weight between the respective users, the second risk score of each user can be calculated according to the similarity weight and the initial score between the target brush popularity user and the target normal user. Specifically, the initial scores of the target brush popularity users B and C are S(B)_0=1S(C)_0=1, and the initial score of the target normal user D is S(D)_0=0, then the suspect A is unknown. The score is calculated as: S(A)_0=(1*0.7+1*0.5)/(0.7+0.5+0.3)=0.75. It should be noted that the score of the user A is only the initial calculated score, and only the connection relationship between the user A and the target brush popularity user and the target normal user is calculated, and the user A and the target brush popularity user are excluded. In the case where the target normal user or other suspected users are connected, in order to improve the accuracy of the calculation, it is necessary to perform multiple rounds of iteration on the algorithm, so that the second risk score of the last A tends to a stable value, and then the stable value is used as the user. A second risk score.

Specifically, if A is connected to user B, C, and D, and is connected to an unknown user E, the similarity weight between user A and user E is first calculated, and assumed to be W(AE)=0.5. , the user A's score is calculated as: S(A)_1=0.8*S(A)_0+0.2*(1*0.7+1*0.5+S(E)_0)*0.5/(0.7+0.5+ 0.3+0.5)

It can be seen that when calculating the score of A in the second round, the score S(A)_0 of the user A calculated in the first round is used as the basic score to participate in the calculation of the second round, and then the iterative calculation of A is performed by the algorithm. Precise user A's second risk score. Preferably, in the embodiment of the present disclosure, when the second risk score of each user is calculated, the number of iterations is 15 times, and the score that tends to be stable after multiple iterations is selected as the second risk score corresponding to the user. The specific algorithm formula of the second risk score is:

Where S _k (i) is the second risk score of user i in the k-th iteration, α is the weight coefficient, which is between 0 and 1, w _ji is the similarity between user j and user i Weights.

Step S160, determining, according to the first risk score and the second risk score, a final risk score of each user as a popular user, and identifying other popular users according to the final risk score.

The user's first risk score is determined by classifying the user, and then the user's second risk score is determined by iteratively calculating the user, and then the user's final risk score is determined according to the first risk score and the second risk score. The final risk score is calculated as:

S _u =w ₁ S _u1 +w ₂ S _u2

Where w _i (i=1, 2) is a weight coefficient, and the value ranges from 0 to 1, and satisfies,

For the first risk score, S _u2 is the second risk score. It is easy to understand that the weighting coefficient W1 is usually smaller than W2 because it is more accurate by the algorithm than by the classification to judge whether the user is a popular user.

Further, comparing the final risk score of each user to the popular user is compared with a predetermined threshold. If the final risk score is greater than the predetermined threshold, determining that the user is a maliciously popular user, and determining whether the user is popular by the means. The user is more accurate, and can estimate whether the user may become a malicious user according to the current final risk score of each user, so as to focus on the subsequent monitoring of the user.

Please refer to FIG. 5 , which is a schematic diagram of functional modules of a device for identifying a popular user of the present disclosure. The device includes an obtaining module 111 , a first determining module 112 , a first identifying module 113 , and a second identifying module 114 . The second determining module 115 and the score determining module 116.

The obtaining module 111 is configured to acquire user characteristics of all users.

In the embodiment of the present disclosure, step S110 may be performed by the acquisition module 111.

The first determining module 112 is configured to determine a first risk score according to a user characteristic corresponding to each user.

In the embodiment of the present disclosure, steps S120-S124 may be performed by the first determining module 112.

The first identification module 113 is configured to identify that the brushed popularity user among all the users is the target brush popularity user.

In the embodiment of the present disclosure, step S130 may be performed by the first identification module 113.

The second identification module 114 is configured to identify the non-brushing popular users determined among all the users as the target normal users.

In the embodiment of the present disclosure, step S140 may be performed by the second identification module 114.

The second determining module 115 is configured to determine a second risk score of each user according to the target brush popularity user and the target normal user.

In the embodiment of the present disclosure, steps S150 to S153 may be performed by the second determination module 115.

The score determination module 116 is configured to determine a final risk score for each user as a popular user based on the first risk score and the second risk score, and identify other popular users based on the final risk score.

In an embodiment of the present disclosure, step S160 may be performed by score determination module 116.

Since it has been described in detail in the method of identifying the popular user, it will not be described here.

In summary, an embodiment of the present disclosure provides a method, an apparatus, a terminal device, and a storage medium for identifying a popular user, and the method and device for identifying the popular user are applied to the terminal device. The method for identifying popular users includes obtaining user characteristics of all users, determining a first risk score according to user characteristics corresponding to each user, and then identifying a popular user determined by all users as a target popular user, and identifying all users. The non-brushing popular user is the target normal user, and the second risk score of each user is determined according to the target brushing popularity user and the target normal user. Finally, the final risk score of each user is determined according to the first risk score and the second risk score. In the present solution, the first risk score and the second risk score of the current user are respectively calculated by two aspects, and then the final risk score of the current user as the popular user is calculated according to the first risk score and the second risk score, and the calculation is performed. More complex, improving the accuracy of identifying users as popular users.

In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may also be implemented in other manners. The apparatus embodiments described above are merely illustrative, for example, the flowcharts and block diagrams in the figures illustrate the architecture, functionality, and functionality of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present disclosure. operating. In this regard, each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that comprises one or more of the Executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may also occur in a different order than those illustrated in the drawings. For example, two consecutive blocks may be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented in a dedicated hardware-based system that performs the specified function or function. Or it can be implemented by a combination of dedicated hardware and computer instructions.

In addition, each functional module in various embodiments of the present disclosure may be integrated to form a separate part, or each module may exist separately, or two or more modules may be integrated to form a separate part.

The functions, if implemented in the form of software functional modules and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, the portion of the technical solution of the present disclosure that contributes in essence or to the prior art or the portion of the technical solution may be embodied in the form of a software product stored in a storage medium, including The instructions are used to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present disclosure. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. . It should be noted that, in this context, relational terms such as first and second are used merely to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply such entities or operations. There is any such actual relationship or order between them. Furthermore, the term "comprises" or "comprises" or "comprises" or any other variations thereof is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device that comprises a plurality of elements includes not only those elements but also Other elements, or elements that are inherent to such a process, method, item, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

The above description is only a preferred embodiment of the present disclosure, and is not intended to limit the disclosure, and various changes and modifications may be made to the present disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and scope of the present disclosure are intended to be included within the scope of the present disclosure. It should be noted that similar reference numerals and letters indicate similar items in the following figures, and therefore, once an item is defined in a drawing, it is not necessary to further define and explain it in the subsequent drawings.

The above is only the specific embodiment of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the disclosure. It should be covered within the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be determined by the scope of the claims.

Industrial applicability

The embodiments of the present disclosure provide a method, a device, a terminal device, and a storage medium for identifying a popular user. The method and device for identifying a popular user are applied to the terminal device, thereby avoiding the omission of the popular user and improving the identification of the user. Brush the accuracy of popular users.

Claims

A method for identifying a popular user, characterized in that the method comprises:

Get user characteristics of all users;

Determining a first risk score according to a user characteristic corresponding to each user;

Identifying the popular users identified by all users as the target to popular users;

Identifying non-brushing popular users identified among all users as target normal users;

Determining a second risk score of each user according to the user characteristics corresponding to the target brush popularity user and the target normal user;

Determining, according to the first risk score and the second risk score, a final risk score of each user as a popular user;

Identify other popular users based on the final risk score.
The method of claim 1, wherein the determining the first risk score according to the user characteristics corresponding to each user comprises:

All users are classified according to the user characteristics by using a predetermined algorithm;

Select key indicators in the user characteristics corresponding to each type of user;

Calculate the average of the key indicators for users under each category;

A first risk score for each user is determined based on the size of the average.
The method of claim 2, wherein the user's key metrics include: the plurality of viewing durations corresponding to the plurality of live rooms being viewed by the user.
The method according to any one of claims 1 to 3, wherein the step of determining the second risk score of each user according to the user characteristics corresponding to the target brush popularity user and the target normal user comprises:

Build a network structure between all users;

Calculating the similarity weight between each two users according to the user characteristics of the target brush popularity user and the target normal user;

Calculating a second risk score of the current user according to a similarity weight of other users connected to the current user in the network structure, and an initial score of the target brush popularity user and the target normal user.
The method of claim 4 wherein said calculating a similarity weight between each two users is:

Among them, Xui is the i-th user feature of user u, Xvi is the i-th user feature of user v, and w uv is the similarity weight between user u and user v.
The method of claim 5, wherein the calculating the second risk score of the current user is:

Where S k (i) is the second risk score of user i in the k-th iteration, α is the weight coefficient, which is between 0 and 1, w ji is the similarity between user j and user i Weights.
The method of any of claims 1-6, wherein the step of identifying other popular users based on the final risk score comprises:

The final risk score for each user is compared to a predetermined threshold, and if the final risk score is greater than a predetermined threshold, the user corresponding to the final risk score is a popular user.
The method of any of claims 1-7, wherein the obtaining user characteristics of all users comprises:

During the live broadcast of the live broadcast, the duration is obtained according to a preset, and the user characteristics of all users in each preset acquisition duration are intermittently obtained at a preset time interval between two consecutive preset acquisition durations. .
The method of any of claims 1-7, wherein the obtaining user characteristics of all users comprises:

During the live broadcast of the live broadcast, and when it is determined that the current popularity value of the live broadcast room is less than a preset popularity threshold, the user characteristics of all users in the live broadcast room are obtained.
An identification device for a popular user, characterized in that the device comprises:

Obtaining a module configured to acquire user characteristics of all users;

a first determining module, configured to determine a first risk score according to a user feature corresponding to each user;

a first identification module configured to identify a popular user identified by all users as a target popular user;

a second identification module configured to identify a non-brushing popular user determined among all users as a target normal user;

a second determining module, configured to determine a second risk score of each user according to the user characteristics corresponding to the target brush popularity user and the target normal user;

The score determining module is configured to determine, according to the first risk score and the second risk score, a final risk score of each user as a popular user; and identify other popular users according to the final risk score.
The apparatus of claim 10, wherein the first determining module is further configured to:

All users are classified according to the user characteristics by using a predetermined algorithm;

Select key indicators in the user characteristics corresponding to each type of user;

Calculate the average of the key metrics for each user under each category;

A first risk score for each user is determined based on the size of the average.
The apparatus according to any one of claims 10 or 11, wherein the second determining module is further configured to:

Build a network structure between all users;

Calculating the similarity weight between each two users according to the user characteristics of the target brush popularity user and the target normal user;

Calculating a second risk score of the current user according to a similarity weight of other users connected to the current user in the network structure, and an initial score of the target brush popularity user and the target normal user.
The device according to any one of claims 10 to 12, wherein the obtaining module is specifically configured to:

During the live broadcast of the live broadcast, the duration is obtained according to a preset, and the user characteristics of all users in each preset acquisition duration are intermittently obtained at a preset time interval between two consecutive preset acquisition durations. .
The device according to any one of claims 10 to 12, wherein the acquiring user characteristics of all users is specifically configured as:

During the live broadcast of the live broadcast, and when it is determined that the current popularity value of the live broadcast room is less than a preset popularity threshold, the user characteristics of all users in the live broadcast room are obtained.
A terminal device, comprising: a memory and a processor, the memory being configured to store computer program code, the processor being configured to execute computer program code stored in the memory to implement a right A method for identifying a popular user of any of claims 1-9.
A readable storage medium, characterized by storing executable instructions that, when executed by one or more processors, implement the method for identifying popular users according to any one of claims 1-9 .