Object recognition method, apparatus, device, and storage medium

Abstract

The present disclosure provides an object identification method and device, equipment and storage medium, relates to the technical field of computers, in particular to the fields of big data, intelligent search and the like. The specific implementation scheme is: according to historical interaction data of a plurality of users, determining a candidate user in the plurality of users; according to the historical interaction data of the candidate user, determining an interaction parameter of the candidate user and a virtual intelligent user; according to the interaction parameter, obtaining interaction data of the virtual intelligent user and the candidate user; according to the interaction data, determining a category of the candidate user, the category of the candidate user indicating whether the candidate user has a preset behavior, and the object being the candidate user having the preset behavior. The identification of the object can be completed for the candidate user suspected of having the preset behavior without affecting other users, and the identification efficiency is also high.

Description

Technical Field

[0001] This disclosure relates to the fields of big data and intelligent search in the field of computer technology, and in particular to an object recognition method, apparatus, device and storage medium. Background Technology

[0002] Quiz games are a popular type of game, and the competitive process between the two sides in the game can increase user engagement.

[0003] In quiz games, some users may use cheats or other methods to win matches. Such behavior undermines the fairness of the game and negatively impacts its normal operation. Therefore, it is crucial to accurately identify users engaging in such unfair practices. Summary of the Invention

[0004] This disclosure provides an object identification method, apparatus, device, and storage medium.

[0005] According to a first aspect of this disclosure, an object recognition method is provided, comprising:

[0006] Based on the historical interaction data of multiple users, candidate users are determined from among the multiple users;

[0007] Based on the historical interaction data of the candidate users, the interaction parameters between the candidate users and the virtual intelligent users are determined;

[0008] The interaction data between the virtual intelligent user and the candidate user is obtained based on the interaction parameters.

[0009] Based on the interaction data, the category of the candidate user is determined. The category of the candidate user indicates whether the candidate user has a preset behavior. The object is the candidate user with the preset behavior.

[0010] According to a second aspect of this disclosure, an object recognition device is provided, comprising:

[0011] The processing unit is used to determine the candidate user from the multiple users based on the historical interaction data of the multiple users;

[0012] The determining unit is used to determine the interaction parameters between the candidate user and the virtual intelligent user based on the candidate user's historical interaction data;

[0013] The acquisition unit is used to acquire the interaction data between the virtual intelligent user and the candidate user based on the interaction parameters;

[0014] The identification unit is used to determine the category of the candidate user based on the interaction data. The category of the candidate user indicates whether the candidate user has a preset behavior, and the object is the candidate user with the preset behavior.

[0015] According to a third aspect of this disclosure, an electronic device is provided, comprising:

[0016] At least one processor; and

[0017] A memory communicatively connected to the at least one processor; wherein,

[0018] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method described in any one of the first aspects.

[0019] According to a fourth aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform the method described in any one of the first aspects.

[0020] According to a fifth aspect of this disclosure, a computer program product is provided, the computer program product comprising: a computer program stored in a readable storage medium, wherein at least one processor of an electronic device can read the computer program from the readable storage medium, and the at least one processor executes the computer program to cause the electronic device to perform the method described in the first aspect.

[0021] The object identification method, apparatus, device, and storage medium disclosed herein first determine candidate users from among multiple users based on their historical interaction data. Then, based on the candidate users' historical interaction data, they determine the interaction parameters between the candidate users and a virtual intelligent user. When the candidate users subsequently interact with other users, they are matched with this virtual intelligent user for interaction. The interaction data between the virtual intelligent user and the candidate users is then obtained based on the interaction parameters, thereby determining the candidate users' category. For candidate users suspected of exhibiting pre-defined behaviors, since a virtual intelligent user is used for matching, the candidate users' interaction data can be obtained without affecting other users, and object identification can be completed based on this interaction data. No manual review is required, resulting in high identification efficiency.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0023] The accompanying drawings are provided to better understand this solution and do not constitute a limitation of this disclosure. Wherein:

[0024] Figure 1 This is a schematic diagram illustrating an application scenario provided by an embodiment of this disclosure;

[0025] Figure 2 A flowchart illustrating the object recognition method provided in this embodiment of the disclosure;

[0026] Figure 3 This is a schematic diagram of the process for determining candidate users provided in an embodiment of the present disclosure;

[0027] Figure 4 This is a schematic diagram illustrating the determination of candidate users provided in an embodiment of the present disclosure;

[0028] Figure 5 A schematic diagram of the process for determining interaction parameters provided in an embodiment of this disclosure;

[0029] Figure 6 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 1 ;

[0030] Figure 7 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 2 ;

[0031] Figure 8 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 3 ;

[0032] Figure 9 This is a schematic diagram of the structure of the object recognition device provided in the embodiments of this disclosure;

[0033] Figure 10 This is a block diagram of an electronic device used to implement the object recognition method of the embodiments of this disclosure. Detailed Implementation

[0034] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0035] Quiz-based competitive games are a popular genre. In these games, the system randomly generates several questions for both players, who then answer them individually. After each round, the system determines the winner based on the players' answers.

[0036] For example, it can be combined Figure 1 The application scenarios of this disclosure are introduced. Figure 1 This is a schematic diagram illustrating an application scenario provided by an embodiment of this disclosure, such as... Figure 1 As shown, User A participates in a quiz game through the first client 11. After starting the quiz game, the system matches User A with an opponent, User B, and then generates corresponding questions. User A can answer the questions through the first client 11, while User B can answer the questions through the second client 12.

[0037] The system can generate one or more questions. For any given question, the time it takes for the question to appear on the first client 11 and the second client 12 is the same. For example... Figure 1 As shown, the question displayed on the first client 11 and the second client 12 is the same: "[When will the bright moon shine on my return?] What is the preceding line?" Below the question, several options are provided. User A can answer the question by touching the screen of the first client 11, and user B can answer the question by touching the screen of the second client 12. The score is related to the user's answering speed and the correctness of the answer. The faster the answering speed and the higher the accuracy, the higher the user's score. The remaining answering time for the current question can also be displayed on the client screen, for example... Figure 1 The first client 11 and the second client 12 display a remaining answering time of 7 seconds. Users A and B must complete the answering within 7 seconds; otherwise, the system will determine the answer to that question as incorrect. Taking a system that generates 10 questions each time as an example, the system will only move to the next question after both users have completed a question or reached the maximum answering time. The winner is determined by the total score across all 10 questions.

[0038] During matches, some users may use cheats or other unfair tactics to win. This behavior negatively impacts other players, making the game unfair and potentially leading to user churn. Therefore, timely identification of users exhibiting such behavior is crucial.

[0039] In this technology, users exhibiting such behavior are primarily identified through a combination of reports from other users and manual review. Specifically, after a match, a report window appears on the game interface. Users can report opponents who they believe have engaged in behavior that undermines game fairness. The reported user is then identified as someone suspected of engaging in unfair play. Upon receiving the report, the backend staff manually review the report to determine if the reported user has indeed engaged in unfair play.

[0040] The method of reporting by other users and then manually reviewing the reports is not very timely. It usually takes several days or even weeks to complete the identification. During this period, the reported user can still play against other users. Such behavior that undermines the fairness of the game is likely to continue for a long time and will also have a negative impact on other users' matches.

[0041] Based on this, this disclosure provides a method for identifying potential users. First, potential users suspected of engaging in behavior detrimental to game fairness are identified from among multiple users. Then, a virtual intelligent user is constructed for each potential user. This virtual intelligent user interacts with the potential user to determine the user's category. Since the interaction is with a virtual intelligent user, it does not affect other users and improves the efficiency of potential user identification. The solution of this disclosure will be described below with reference to the accompanying drawings.

[0042] Figure 2 This is a flowchart illustrating the object recognition method provided in the embodiments of this disclosure, such as... Figure 2 As shown, the method may include:

[0043] S21, Based on the historical interaction data of multiple users, determine the candidate users from among the multiple users.

[0044] Users can interact with other users on the corresponding platform using their user accounts. Taking a game platform as an example, multiple users refer to those who have played against each other in the game. For any given interaction, the server can obtain the corresponding interaction data. In this embodiment, the historical interaction data refers to the interaction data corresponding to interactions completed by multiple users. For example, historical interaction data could be interaction data from the most recent week, the most recent month, the most recent three days, and so on.

[0045] Historical interaction data includes the interaction content between two users. Taking a quiz game platform as an example, historical interaction data may include data such as the user's answering status and reporting status during the battle. Answering status may include data such as the time it takes for the user to answer questions and the accuracy rate of the answers. Reporting status may include data such as whether the user reported the user they were playing against or whether they were reported by other users. Historical interaction data may also include the accounts of the two users who are interacting.

[0046] Candidate users are those suspected of exhibiting pre-defined behaviors. In the game, pre-defined behaviors include various actions that undermine the fairness of the game, such as using cheats or hacks. After obtaining historical interaction data from multiple users, the server can determine candidate users from among them based on this data. Candidate users are those whose historical interaction data differs from that of normal users. For example, normal users receive fewer reports, while candidate users receive more. Or, normal users' accuracy or speed in answering questions is within a certain range, while candidate users' accuracy or speed is abnormal. Based on this historical interaction data, candidate users can be determined from among multiple users.

[0047] S22, determine the interaction parameters between the candidate user and the virtual intelligent user based on the candidate user's historical interaction data.

[0048] The candidate users are those suspected of having pre-defined behaviors; that is, the server has not yet determined whether the candidate user has pre-defined behaviors and needs to make further determinations in subsequent processes. Once the candidate users are identified, when they enter the corresponding platform, they will not be matched with other users. Instead, a virtual intelligent user will be created to interact with the candidate user.

[0049] The interaction parameters of the virtual intelligent user are determined based on the historical interaction data of the candidate users; that is, different candidate users correspond to different virtual intelligent users. Since the historical interaction data can identify candidate users who are suspected of exhibiting pre-defined behaviors, the interaction parameters determined based on the candidate users' historical interaction data are also related to the pre-defined behaviors that the candidate users are suspected of exhibiting. For example, if the candidate users have a high accuracy rate in answering questions, the interaction parameters can be set to ensure that the virtual intelligent user's accuracy rate is higher than that of the candidate users; similarly, if the candidate users answer questions quickly, the interaction parameters can be set to ensure that the virtual intelligent user's answering speed is faster than that of the candidate users, and so on.

[0050] S23, obtain the interaction data between the virtual intelligent user and the candidate user based on the interaction parameters.

[0051] After determining the interaction parameters of the candidate user and the virtual AI user, the system will then match the virtual AI user with the candidate user for interaction. For example, in a gaming platform, when the candidate user enters the game, the virtual AI user will be their opponent. The candidate user and the virtual AI user will answer questions together, and the server will then obtain the corresponding interaction data. This interaction data may include, for example, the candidate user's answering speed and accuracy rate during the interaction with the virtual AI user.

[0052] S24. Based on the interaction data, determine the category of the candidate user. The category of the candidate user indicates whether the candidate user has a preset behavior. The object is the candidate user with the preset behavior.

[0053] Since virtual intelligent users interact with candidate users based on interaction parameters, which are determined based on the candidate users' historical interaction data, virtual intelligent users can, to some extent, stimulate users with preset behaviors to further perform preset behaviors.

[0054] For example, if a candidate user answers questions quickly, the virtual AI user can be set to answer even faster. If the candidate user is using game cheats to increase their speed, and the virtual AI user is also answering quickly, the candidate user might modify the cheat parameters to further increase their speed in order to win the match. In this case, the candidate user can be identified as having pre-defined behavior based on their answering speed.

[0055] If the candidate user is not using game cheats, even if the virtual AI user answers faster, the candidate user's answering speed will not change significantly. In this case, the candidate user can be determined to be someone without pre-defined behavior based on their answering speed.

[0056] In this embodiment of the disclosure, the object is a candidate user with preset behaviors. After determining the category of the candidate user through interaction data, the object can be identified based on the category of the candidate user, that is, the candidate user with preset behaviors.

[0057] The object identification method provided in this disclosure first identifies candidate users from among multiple users based on their historical interaction data. Then, it determines the interaction parameters between the candidate users and a virtual intelligent user based on the candidate users' historical interaction data. When the candidate users subsequently interact with other users, they are matched with the virtual intelligent user for interaction. The interaction data between the virtual intelligent user and the candidate users is then obtained based on the interaction parameters, thereby determining the category of the candidate users. For candidate users suspected of exhibiting pre-defined behaviors, since a virtual intelligent user is used for matching, the candidate users' interaction data can be obtained without affecting other users, and object identification can be completed based on this interaction data. No manual review is required, resulting in high identification efficiency.

[0058] To help readers gain a deeper understanding of the implementation principles of this disclosure, the following will be discussed in conjunction with... Figures 3-8 right Figure 2 The illustrated embodiments are further refined.

[0059] First, combine Figure 3 The process of determining the candidates is described. Figure 3 This is a schematic diagram of the process for determining candidate users provided in an embodiment of this disclosure, such as... Figure 3 As shown, it includes:

[0060] S31. Based on the historical interaction data of multiple users, obtain the indicator data of multiple users. The indicator data includes at least one of the following: answering speed, answering accuracy rate, or number of reports.

[0061] After a user participates in an interaction, the system can obtain the user's historical interaction data. In this embodiment of the disclosure, the historical interaction data may include user interaction behavior logs, or user report data.

[0062] Users can be rated based on their behavior logs and complaint data. For example, one possible user rating would be as follows:

[0063] L1: High-quality users refer to users who have no data anomalies and have not been reported by other users;

[0064] L2: Benign users refer to users who do not have data anomalies and whose reports from other users are less than a certain threshold;

[0065] L3: Candidate users refer to users with minor data anomalies or whose number of reports is greater than or equal to a certain threshold;

[0066] L4: Abnormal users refer to users with serious data anomalies, who need to be banned for a period of time.

[0067] L5: Users with serious abnormalities require permanent banning.

[0068] Based on user behavior logs, users at levels L4 and L5, i.e., users with serious data anomalies, can be identified first. The following will use a quiz game as an example to introduce this process.

[0069] Behavior logs can include the time when questions appear in a quiz game, the time when a user taps the client's screen, and the location of those taps. For example, if user A and user B participate in an interaction, the system generates 10 questions for them. For any given question, the system records the time it appears on the client's screen; the time it appears on user A's screen is the same as the time it appears on user B's screen. When user A or user B answers a question, they will perform a tap on the client's screen, and the system will also record the time and location of these taps.

[0070] After obtaining the behavior logs of multiple users, users with serious data anomalies can be filtered out based on these logs. Users with serious data anomalies refer to those who engage in behaviors that severely undermine the fairness of the game; these behaviors are those that normal users would not be able to perform. Examples include completing a question before it appears or completing a question without clicking on the client screen, etc.

[0071] Completing an answer before the question appears constitutes premature answering, which can be determined based on behavior logs. Since behavior logs record the time each question is displayed on the client screen and the time the user clicks the screen, for any given question, if the question is displayed on the client screen at time A, and the behavior log records the user clicking the screen to complete the question at time B, and time B is earlier than time A, then it can be determined that the user has engaged in premature answering.

[0072] The act of completing a question without clicking the client screen can also be determined based on behavior logs. Since every time a user clicks the client screen to answer a question, the backend records it in the behavior log. For any given question, if the user does not click the client screen but the backend determines that the user has completed the question, then it can be determined that the user has completed the question without clicking the client screen.

[0073] Actions such as answering questions prematurely or completing answers without clicking the client screen constitute serious unfairness in the game. These behaviors typically occur due to the use of game cheats or similar methods; normal users would not engage in such behavior. Therefore, when such seriously unfair behavior is identified based on behavior logs, the corresponding user can be classified as having pre-defined behaviors. For example, based on the user rating system described in the above embodiment, such seriously unfair behavior can be characterized as severe data anomalies, and users exhibiting such behaviors can be classified as L4 or L5 level users.

[0074] For users other than those who have engaged in behavior that seriously undermines the fairness of the game, user metrics data can be obtained based on the user's historical interaction data. These metrics data include at least one of the following: answering speed, answering accuracy rate, or number of reports.

[0075] Since the behavior log records the time each question appears on the client screen and the time the user taps the screen, the user's answering speed for that question can be determined based on these times. For example, if a question appears on the client screen at 8:23:00 and the user completes the answer at 8:23:05, then the user took 5 seconds to complete the question, and their answering speed is 1 / 5 question / second.

[0076] Because the quiz game has a maximum answering time limit, the behavior log may record instances where, for example, if neither user answers within 10 seconds (maximum answering time is 10 seconds), the system automatically moves on to the next question. In this case, the behavior log will not include the time the user took to complete that question. To address this, the time the user takes to complete that question can be defined as the maximum answering time. Using 10 seconds as an example, the user's answering speed would be 1 / 10 of a question per second.

[0077] The indicator data in this embodiment includes answering speed, which can be the speed at which a user completes a single question or the speed at which a user completes multiple questions in a single battle. Taking a battle consisting of 5 questions as an example, if a user takes 3 seconds, 5 seconds, 6 seconds, 2 seconds, and 8 seconds to complete these 5 questions, respectively, then the user's answering speed for a single question is 1 / 3 question / second, 1 / 5 question / second, 1 / 6 question / second, 1 / 2 question / second, and 1 / 8 question / second. The user's answering speed for multiple questions in this battle is 5 / (3+5+6+2+8) = 1 / 4.8 question / second. Therefore, for this battle, 1 / 3 question / second, 1 / 5 question / second, 1 / 6 question / second, 1 / 2 question / second, and 1 / 8 question / second can all be considered as the user's answering speed, or 1 / 4.8 seconds per question can be considered as the user's answering speed.

[0078] Since the behavior log records the user's tap locations on the client screen, it's possible to determine whether the user answered a question correctly based on these tap locations and the positions of the question's options on the client screen. If the behavior log doesn't include the user's tap locations for a particular question, it means the user didn't answer that question, and in this case, it can be assumed the user answered incorrectly. For example, in a battle with 10 questions, if a user answers 7 correctly, their accuracy rate for that battle is 70%.

[0079] The above embodiments describe a method for obtaining a user's answering speed and answering accuracy based on behavior logs. In some embodiments, historical interaction data may also include user report data, and the number of user reports can be obtained based on the user report data.

[0080] For example, for user A, the users who are playing against user A include user B, user C, user D, etc. Among them, user B and user D reported user A, so user A's report count is 2.

[0081] S32, based on the indicator data of multiple users, determine the candidate users among the multiple users. The candidate users are those whose corresponding indicator data is within the preset range.

[0082] After obtaining the metrics data of multiple users, candidate users can be determined based on the metrics data, which include at least one of the following: answering speed, answering accuracy, and number of reports.

[0083] In one possible implementation, users whose answering speed is less than or equal to a first threshold can be identified as candidate users. This first threshold can be preset. For example, if the first threshold is 1 / 8 of a question per second, a speed less than or equal to 1 / 8 of a question per second indicates that the user's answering speed is relatively slow, potentially indicating passive answering behavior. Therefore, users suspected of passive answering behavior are identified as candidate users.

[0084] In one possible implementation, users whose answering speed is greater than or equal to a second threshold can be identified as candidate users. The second threshold can be preset and is greater than or equal to a first threshold. For example, if the second threshold is 1 question / second, an answering speed greater than or equal to 1 question / second indicates that the user's answering speed is relatively fast, potentially suggesting the use of cheats or other methods to increase their answering speed. Therefore, users suspected of using cheats or other methods to increase their answering speed are identified as candidate users.

[0085] In one possible implementation, users with a correct answer rate greater than or equal to a third threshold can be identified as candidate users. This third threshold can be preset and is a value between 0 and 1. For example, with a third threshold of 0.95, a user's correct answer rate greater than or equal to 0.95 indicates a relatively high accuracy rate, potentially suggesting the user may have used cheats or other methods to improve their accuracy. Therefore, users with a correct answer rate greater than or equal to the third threshold are identified as candidate users.

[0086] In one possible implementation, users with a report count greater than or equal to a fourth threshold can be identified as candidate users. This fourth threshold can be preset and is a positive integer greater than or equal to 1. For any given user, if another user reports that user after they have played against them, that user's report count will increase by one. The user's report count can be obtained based on the other users' report data. When a user's report count is greater than or equal to the fourth threshold, it indicates that the user has been reported by a large number of other users, and there is a high probability that the user has engaged in pre-defined behavior. In this case, the user can be identified as a candidate user.

[0087] Figure 4 This is a schematic diagram illustrating the determination of candidate users provided in an embodiment of this disclosure, such as... Figure 4 As shown, five users are illustrated: User 41, User 42, User 43, User 44, and User 45. The metrics for these five users are as follows:

[0088] User 41 - Answering speed: 2 questions / second; Correct answer rate: 75%; Number of reports: 3;

[0089] User 42 - Answering speed: 1 / 3 question / second; Answering accuracy: 70%; Number of reports: 0.

[0090] User 43 - Answering speed: 1 / 10 questions / second, accuracy rate: 10%, number of reports: 23;

[0091] User 44 - Answering speed: 1 / 5 questions / second; Answering accuracy: 100%; Number of reports: 6;

[0092] User 45 - Answering speed 1 / 3 of a question per second, accuracy rate 85%, number of reports 36.

[0093] The first threshold is 1 / 8, the second threshold is 1 / 2, the third threshold is 99%, and the fourth threshold is 20. For user 41, their answer accuracy is less than the third threshold, their number of reports is less than the fourth threshold, but their answering speed is greater than or equal to the second threshold. Therefore, user 41 is selected as a candidate user. For user 42, their answering speed is greater than the first threshold and less than the second threshold, their answer accuracy is less than the third threshold, and their number of reports is less than the fourth threshold. Therefore, user 42 is not a candidate user. For user 43, their answer accuracy is less than the third threshold, but their answering speed is less than the first threshold, and their number of reports is greater than the fourth threshold. Therefore, user 43 is selected as a candidate user. For user 44, their answering speed is greater than the first threshold and less than the second threshold, their number of reports is less than the fourth threshold, but their answer accuracy is greater than the third threshold. Therefore, user 44 is selected as a candidate user. For user 45, their answering speed is greater than the first threshold and less than the second threshold, their answer accuracy is less than the third threshold, but their number of reports is greater than the fourth threshold. Therefore, user 45 is selected as a candidate user.

[0094] Therefore, in Figure 4 In the example, based on the metrics data of 5 users, users 41, 43, 44 and 45 can be identified as candidate users with a user rating of L3, while user 42 is not a candidate user and has a user rating of L1.

[0095] The above embodiments describe a method for determining candidate users. By using metrics such as answering speed, answering accuracy, and the number of reports, candidate users can be identified relatively accurately from multiple users. This can, to some extent, reduce the impact of candidate users on other users and improve their experience. After determining the candidate users from multiple users, it is necessary to construct a virtual intelligent user for them and interact with them. This process will be described below.

[0096] For example, it can be combined Figure 5 This section describes the process of building virtual intelligent users. Figure 5 This is a schematic diagram of the process for determining interaction parameters provided in an embodiment of this disclosure, such as... Figure 5 As shown, it includes:

[0097] S51, based on the historical interaction data of the candidate users, determine at least one of the candidate users' average answering speed and average answering accuracy during the historical period.

[0098] After the candidate users are determined, their historical interaction data includes their behavior logs within a historical time period. Based on these logs, the time and location of each user's clicks on the client screen during game battles can be obtained. The duration of the user's completion of a question can be determined by the time the user clicks on the client screen and the time the question appears on the screen. Finally, the location of the user's clicks on the client screen can be used to determine whether the user answered the question correctly.

[0099] Then, based on the time taken by the user to complete each question and the number of questions, the average answering speed of the candidate user in the historical period can be obtained, and based on the results of the user completing each question, the average answering accuracy of the candidate user in the historical period can be obtained.

[0100] S52, obtain the interaction parameters based on at least one of the average answering time and average answering accuracy.

[0101] Since the candidate users are suspected of having pre-set behaviors, that is, suspected of undermining the fairness of the game, it is impossible to determine whether the candidate users have pre-set behaviors based on their historical interaction data. Therefore, further judgment is needed for the candidate users.

[0102] The candidate users may have pre-defined behaviors. In order to reduce the impact of such pre-defined behaviors on other users, a virtual intelligent user will be built for the candidate users to interact with.

[0103] Virtual intelligent users are constructed based on the historical interaction data of candidate users. The attributes of a virtual intelligent user are determined by its interaction parameters, which may include, for example, the user's answering speed and accuracy. The interaction parameters of the constructed virtual intelligent user will differ for different candidate users.

[0104] For candidate users whose average answering speed is less than or equal to a first threshold, the answering speed of the virtual intelligent user can be obtained based on the average answering speed of the candidate users, wherein the answering speed of the virtual intelligent user is less than or equal to the average answering speed of the candidate users.

[0105] For candidate users whose average answering speed is greater than or equal to the second threshold, the answering speed of the virtual intelligent user can be obtained based on the average answering speed of the candidate users, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed of the candidate users.

[0106] For candidate users whose average answer accuracy is greater than or equal to the third threshold, the answering speed and answering accuracy of virtual intelligent users can be obtained based on the average answer accuracy and average answering speed of candidate users. Specifically, the answering speed of virtual intelligent users is greater than or equal to the average answering speed of candidate users, and the answering accuracy of virtual intelligent users is greater than or equal to the average answering accuracy of candidate users.

[0107] After constructing the virtual intelligent user, the system configures this virtual intelligent user to interact with the candidate user. Then, the system acquires interaction data between the candidate user and the virtual intelligent user, where the virtual intelligent user interacts with the candidate user according to interaction parameters. Based on the interaction data, the system can obtain the candidate user's initial answering speed and initial answering accuracy during the interaction with the virtual intelligent user.

[0108] Since the interaction parameters of virtual intelligent users are determined based on at least one of the candidate users' average answering speed and average answering accuracy, the behavior of virtual intelligent users will further stimulate the behavior of abnormal users, while having little impact on the behavior of normal users. Constructing virtual intelligent users can improve the accuracy of candidate user identification. Therefore, the category of candidate users can be determined based on at least one of average answering speed, average answering accuracy, first answering speed, and first answering accuracy.

[0109] The following section will explain how to determine the category of the candidate user, with reference to the attached diagram.

[0110] Figure 6 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 1 ,like Figure 6As shown, the candidate user is User A 61, who will compete against the virtual intelligent user 62. User A 61 is a candidate user whose average answering speed is less than or equal to a first threshold. Figure 6 In the test, user A61's average answering speed was 1 / 8 of a question per second, and the first threshold was 1 / 8.

[0111] When the average answering speed of the candidate users is less than or equal to the first threshold, the answering speed of the virtual intelligent user will be set to be less than or equal to the average answering speed. Figure 6 In the example, the virtual AI user 62 answers questions at a speed of 1 / 9 of a question per second. This means that if the candidate user's answering speed is too low, the virtual AI user's answering speed can be set to be lower than or equal to that of the candidate user.

[0112] Then, based on the interaction data between the candidate user and the virtual AI user, the initial answering speed of the candidate user during their interaction with the virtual AI user is obtained. Figure 6 The example shows interaction data 63 between User A 61 and Virtual Intelligent User 62, in which User A 61 and Virtual Intelligent User 62 answered five questions. Figure 6 The example shows the time it takes for the five questions to be displayed on the client screen, as well as the time it takes for User A 61 and Virtual Intelligent User 62 to complete the five questions respectively. Based on the current time on the client screen for the five questions and the time it takes for User A 61 to complete the five questions, we can obtain the total time spent by User A 61 to complete the five questions: t = 3 + 5 + 9 + 8 + 6 = 31 seconds. Therefore, the first answering speed v = 5 / 31 questions / second. Since the first answering speed is greater than the average answering speed of 1 / 8 questions / second, the candidate user (i.e., User A 61) is determined to be a user without preset behavior.

[0113] Figure 7 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 2 ,like Figure 7 As shown, the candidate user is User B 71, who will compete against the virtual intelligent user 72. User B 71 is a candidate user whose average answering speed is greater than or equal to the second threshold. Figure 7 In the test, user B71's average answering speed was 1 question / second, and the second threshold was 1 / 2.

[0114] When the average answering speed of the candidate users is greater than or equal to the second threshold, the answering speed of the virtual intelligent user will be set to be greater than or equal to the average answering speed. Figure 7In the example, the virtual AI user 72 answers 1 question per second. This means that if the candidate user's answering speed is too fast, the virtual AI user's answering speed can be set to be faster than or equal to that of the candidate user. Since the virtual AI user's answering speed is also relatively fast, if the candidate user has pre-defined behaviors, they might further increase their answering speed to win the game. If the candidate user does not have pre-defined behaviors, their answering speed should remain roughly the same when interacting with the virtual AI user. Based on this, the candidate user can be identified.

[0115] Specifically, based on the interaction data between the candidate user and the virtual AI user, the initial answering speed of the candidate user during their interaction with the virtual AI user is obtained. Figure 7 The example shows interaction data 73 between user B 71 and virtual intelligent user 72, in which user B 71 and virtual intelligent user 72 competed in a five-question game. Figure 7 The example shows the time it takes for these five questions to be displayed on the client screen, as well as the time it takes for User B 71 and Virtual Intelligent User 72 to complete these five questions respectively. Based on the current time on the client screen for the five questions and the time it takes for User B 71 to complete these five questions, we can obtain the total time spent by User B 71 to complete these five questions: t = 1 + 0.8 + 0.7 + 0.2 + 0.6 = 3.3 seconds. Therefore, the first answering speed v = 5 / 3.3 questions / second. Since the first answering speed is greater than the average answering speed of 1 question / second, the candidate user (i.e., User B 71) is determined to be a user with the preset behavior.

[0116] Figure 8 An illustration of determining the category of candidate users provided in embodiments of this disclosure. Figure 3 ,like Figure 8 As shown, the candidate user is user C81, who interacts with virtual intelligent user 82. User C81 is a candidate user whose average answer accuracy is greater than or equal to the third threshold. User C81's average answering speed is 1 / 5 question / second, the average answer accuracy is 99%, and the third threshold is 98%.

[0117] When the average correct answer rate of the candidate users is greater than or equal to the third threshold, the correct answer rate of the virtual intelligent user will be set to be greater than or equal to the average correct answer rate, and the answering speed of the virtual intelligent user will be greater than or equal to the average answering speed. Figure 8 In the example, the virtual AI user 82 answers questions at a speed of 1 / 4 question per second, with an accuracy rate of 99%. That is, when the candidate user's accuracy rate is too high, the virtual AI user's accuracy rate can be set to be equal to or higher than the candidate user's accuracy rate, and the virtual AI user's answering speed should not be lower than the candidate user's answering speed.

[0118] Because virtual AI users answer questions quickly and accurately, if the candidate user exhibits pre-defined behaviors, they might further increase their answering speed or accuracy to win the match. Conversely, if the candidate user does not exhibit pre-defined behaviors, their answering speed and accuracy should remain roughly the same when competing against the virtual AI user. Based on this, candidate user identification can be achieved.

[0119] Specifically, based on the interaction data between the candidate user and the virtual AI user, the initial answering speed and initial answering accuracy of the candidate user during their interaction with the virtual AI user are obtained. Figure 8 The example 83 shows the interaction data 83 between user C81 and virtual intelligent user 82, in which user C81 and virtual intelligent user 82 answered five questions. Figure 8 The example shows the time it takes for the five questions to appear on the client screen, as well as the time and accuracy of user C81 and virtual intelligent user 82 in completing these five questions. Based on the current time on the client screen for the five questions and the time it takes user C81 to complete them, the total time user C81 spent completing the five questions is t = 3 + 2 + 2 + 3 + 3 = 13 seconds. Therefore, the first answering speed is v = 5 / 13 questions / second. Since user C81 answered all five questions correctly, their first answering accuracy is 100%.

[0120] Since the first correct answer rate is greater than or equal to the average correct answer rate, and the first answer speed is greater than or equal to the average answer speed, the candidate user (i.e., user C81) is determined to be a user with preset behavior.

[0121] In summary, this disclosure provides a method for identifying potential users. First, it identifies potential users with pre-defined behaviors among the existing user base. Then, it determines the interaction parameters of a virtual intelligent user based on the historical interaction data of these potential users, thus constructing the virtual intelligent user. The virtual intelligent user interacts with the potential users according to these interaction parameters, without affecting other users. Since the interaction parameters of the virtual intelligent user are set based on the historical interaction data of the potential users, the virtual intelligent user can further stimulate users with pre-defined behaviors to execute those behaviors. Therefore, after a potential user and a virtual intelligent user engage in combat, the interaction data between the virtual intelligent user and the potential user can be used to determine whether the potential user has further executed the pre-defined behavior, thereby determining the category of the potential user. This disclosed solution eliminates the need for manual review, has good timeliness, and improves the efficiency of potential user identification.

[0122] Figure 9 This is a schematic diagram of the structure of the object recognition device provided in the embodiments of this disclosure, such as... Figure 9 As shown, the candidate user identification device 90 includes:

[0123] Processing unit 91 is used to determine candidate users from among the multiple users based on the historical interaction data of multiple users;

[0124] The determining unit 92 is used to determine the interaction parameters between the candidate user and the virtual intelligent user based on the candidate user's historical interaction data;

[0125] Acquisition unit 93 is used to acquire interaction data between the virtual intelligent user and the candidate user based on the interaction parameters;

[0126] The identification unit 94 is used to determine the category of the candidate user based on the interaction data. The category of the candidate user indicates whether the candidate user has a preset behavior. The object is the candidate user with the preset behavior.

[0127] In one possible implementation, the processing unit 91 includes:

[0128] The first acquisition module is used to acquire indicator data of the multiple users based on their historical interaction data. The indicator data includes at least one of the following: answering speed, answering accuracy, or number of reports.

[0129] The first determining module is used to determine the candidate user from among the multiple users based on the indicator data of the multiple users, wherein the candidate user is a user whose corresponding indicator data is within a preset range.

[0130] In one possible implementation, the first determining module includes:

[0131] The determination submodule is used to determine that the candidate users include at least one of the following:

[0132] Users whose answering speed is less than or equal to the first threshold;

[0133] Users whose answering speed is greater than or equal to the second threshold;

[0134] Users whose answer accuracy rate is greater than or equal to the third threshold;

[0135] Users whose number of reports is greater than or equal to the fourth threshold.

[0136] In one possible implementation, the determining unit 92 includes:

[0137] The second determining module is used to determine, based on the candidate user's historical interaction data, at least one of the candidate user's average answering speed and average answering accuracy during the historical time period;

[0138] The second acquisition module is used to acquire the interaction parameters based on at least one of the average answering time and the average answering accuracy.

[0139] In one possible implementation, the second acquisition module includes:

[0140] The first acquisition submodule is used to acquire the answering speed of the virtual intelligent user based on the average answering speed when the average answering speed is less than or equal to the first threshold, wherein the answering speed of the virtual intelligent user is less than or equal to the average answering speed;

[0141] The second acquisition submodule is used to acquire the answering speed of the virtual intelligent user based on the average answering speed when the average answering speed is greater than or equal to the second threshold, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed;

[0142] The third acquisition submodule is used to acquire the answering speed and answering accuracy of the virtual intelligent user based on the average answering speed and the average answering accuracy when the average answering accuracy is greater than or equal to the third threshold, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed, and the answering accuracy of the virtual intelligent user is greater than or equal to the average answering accuracy.

[0143] In one possible implementation, the identification unit 94 includes:

[0144] The third acquisition module is used to acquire, based on the interaction data, at least one of the first answering speed and the first answering accuracy rate of the candidate user interacting with the virtual intelligent user;

[0145] The identification module is used to determine the category of the candidate user based on at least one of the average answering speed, the average answering accuracy, the first answering speed, and the first answering accuracy.

[0146] In one possible implementation, the identification module includes:

[0147] The first identification submodule is used to determine that the candidate user has the preset behavior if the first answering speed is less than or equal to the average answering speed when the average answering speed is less than or equal to the first threshold.

[0148] The second identification submodule is used to determine that the candidate user is a user with the preset behavior if the first answering speed is greater than or equal to the average answering speed when the average answering speed is greater than or equal to the second threshold.

[0149] The third identification submodule is used to determine that the candidate user is a user with the preset behavior when the average answer accuracy rate is greater than or equal to the third threshold, and the first answer accuracy rate is greater than or equal to the average answer accuracy rate and the first answer speed is greater than or equal to the average answer speed.

[0150] The fourth identification submodule is used to determine that the candidate user is a user who does not have the preset behavior.

[0151] The object recognition device provided in this embodiment is used to execute the above method embodiment. Its implementation principle and technical effect are similar, and will not be described again here.

[0152] This disclosure provides an object recognition method, apparatus, device, and storage medium, which are applied in the fields of big data and intelligent search in the field of computer technology, so as to accurately identify users who engage in behaviors that undermine the fairness of the game, thereby improving the fairness of the game.

[0153] The collection, storage, use, processing, transmission, provision, and disclosure of user personal information involved in the technical solution disclosed herein comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0154] According to embodiments of this disclosure, this disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0155] According to embodiments of this disclosure, this disclosure also provides a computer program product comprising: a computer program stored in a readable storage medium, at least one processor of an electronic device being able to read the computer program from the readable storage medium, and the at least one processor executing the computer program causing the electronic device to perform the scheme provided in any of the above embodiments.

[0156] Figure 10 A schematic block diagram of an example electronic device 1000 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0157] like Figure 10As shown, device 1000 includes a computing unit 1001, which can perform various appropriate actions and processes according to a computer program stored in read-only memory (ROM) 1002 or a computer program loaded from storage unit 1008 into random access memory (RAM) 1003. The RAM 1003 may also store various programs and data required for the operation of device 1000. The computing unit 1001, ROM 1002, and RAM 1003 are interconnected via bus 1004. Input / output (I / O) interface 1005 is also connected to bus 1004.

[0158] Multiple components in device 1000 are connected to I / O interface 1005, including: input unit 1006, such as keyboard, mouse, etc.; output unit 1007, such as various types of monitors, speakers, etc.; storage unit 1008, such as disk, optical disk, etc.; and communication unit 1009, such as network card, modem, wireless transceiver, etc. Communication unit 1009 allows device 1000 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0159] The computing unit 1001 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1001 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1001 performs the various methods and processes described above, such as object recognition methods. For example, in some embodiments, the object recognition method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 1008. In some embodiments, part or all of the computer program may be loaded and / or installed on device 1000 via ROM 1002 and / or communication unit 1009. When the computer program is loaded into RAM 1003 and executed by the computing unit 1001, one or more steps of the object recognition method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to perform object recognition methods by any other suitable means (e.g., by means of firmware).

[0160] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0161] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0162] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0163] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device for displaying information to the user (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor); and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0164] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0165] Computer systems can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. A server can be a cloud server, also known as a cloud computing server or cloud host, a hosting product within the cloud computing service ecosystem, addressing the shortcomings of traditional physical hosts and VPS (Virtual Private Server, or simply "VPS") services, such as high management difficulty and weak business scalability. Servers can also be servers for distributed systems or servers incorporating blockchain technology.

[0166] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0167] The specific embodiments described above do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A method for object identification, comprising: determining a candidate user from a plurality of users according to historical interaction data of the plurality of users; determining at least one of an average answering speed and an average answering accuracy of the candidate user within a historical period according to historical interaction data of the candidate user; obtaining an interaction parameter of the candidate user and a virtual intelligent user according to at least one of the average answering speed and the average answering accuracy; obtaining interaction data of the virtual intelligent user and the candidate user according to the interaction parameter; determining a category of the candidate user according to the interaction data, the category of the candidate user indicating whether the candidate user has a preset behavior, the object being the candidate user having the preset behavior; obtaining an interaction parameter of the candidate user and a virtual intelligent user according to at least one of the average answering speed and the average answering accuracy, comprising: when the average answering speed is less than or equal to a first threshold, obtaining an answering speed of the virtual intelligent user according to the average answering speed, wherein the answering speed of the virtual intelligent user is less than or equal to the average answering speed; when the average answering speed is greater than or equal to a second threshold, obtaining an answering speed of the virtual intelligent user according to the average answering speed, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed; when the average answering accuracy is greater than or equal to a third threshold, obtaining an answering speed and an answering accuracy of the virtual intelligent user according to the average answering speed and the average answering accuracy, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed, and the answering accuracy of the virtual intelligent user is greater than or equal to the average answering accuracy.

2. The method of claim 1, wherein, determining a candidate user from a plurality of users according to historical interaction data of the plurality of users, comprising: obtaining index data of the plurality of users according to historical interaction data of the plurality of users, the index data comprising at least one of an answering speed, an answering accuracy, or a reporting amount; determining the candidate user from the plurality of users according to the index data of the plurality of users, wherein the candidate user is a user whose corresponding index data is within a preset range.

3. The method of claim 2, wherein, determining the candidate user from the plurality of users according to the index data of the plurality of users, comprising: determining the candidate user to comprise at least one of: a user whose answering speed is less than or equal to the first threshold; a user whose answering speed is greater than or equal to the second threshold; a user whose answering accuracy is greater than or equal to the third threshold; a user whose reporting amount is greater than or equal to a fourth threshold.

4. The method of claim 3, wherein, determining a category of the candidate user according to the interaction data, comprising: obtaining at least one of a first answering speed and a first answering accuracy of the candidate user and the virtual intelligent user according to the interaction data; determining the category of the candidate user according to at least one of the average answering speed, the average answering accuracy, the first answering speed, and the first answering accuracy.

5. The method of claim 4, wherein, According to at least one of the average answering speed, the average answering accuracy, the first answering speed and the first answering accuracy, a category of the candidate user is determined, comprising: When the average answering speed is less than or equal to the first threshold, if the first answering speed is less than or equal to the average answering speed, the candidate user is determined as a user with the preset behavior; When the average answering speed is greater than or equal to the second threshold, if the first answering speed is greater than or equal to the average answering speed, the candidate user is determined as a user with the preset behavior; When the average answering accuracy is greater than or equal to the third threshold, if the first answering accuracy is greater than or equal to the average answering accuracy, and the first answering speed is greater than or equal to the average answering speed, the candidate user is determined as a user with the preset behavior; Otherwise, the candidate user is determined as a user without the preset behavior.

6. An object recognition apparatus, comprising: a processing unit configured to determine a candidate user from a plurality of users according to historical interaction data of the plurality of users; a determination unit configured to determine an interaction parameter of the candidate user with a virtual intelligent user according to historical interaction data of the candidate user; an acquisition unit configured to acquire interaction data of the virtual intelligent user with the candidate user according to the interaction parameter; a recognition unit configured to determine a category of the candidate user according to the interaction data, the category of the candidate user indicating whether the candidate user has a preset behavior, and the object being the candidate user with the preset behavior; the determination unit comprises: a second determination module configured to determine at least one of an average answering speed and an average answering accuracy of the candidate user in a historical period according to the historical interaction data of the candidate user; a second acquisition module configured to acquire the interaction parameter according to at least one of the average answering speed and the average answering accuracy; the second acquisition module comprises: a first acquisition sub-module configured to acquire an answering speed of the virtual intelligent user according to the average answering speed when the average answering speed is less than or equal to a first threshold, wherein the answering speed of the virtual intelligent user is less than or equal to the average answering speed; a second acquisition sub-module configured to acquire an answering speed of the virtual intelligent user according to the average answering speed when the average answering speed is greater than or equal to a second threshold, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed; a third acquisition sub-module configured to acquire an answering speed and an answering accuracy of the virtual intelligent user according to the average answering speed and the average answering accuracy when the average answering accuracy is greater than or equal to a third threshold, wherein the answering speed of the virtual intelligent user is greater than or equal to the average answering speed, and the answering accuracy of the virtual intelligent user is greater than or equal to the average answering accuracy.

7. The apparatus of claim 6, wherein, the processing unit comprises: The first obtaining module is configured to obtain index data of the plurality of users according to historical interaction data of the plurality of users, the index data comprising at least one of: an answering speed, an answering accuracy, or a reporting amount. The first determining module is configured to determine the candidate user from the plurality of users according to the index data of the plurality of users, wherein the candidate user is a user whose corresponding index data is within a preset range.

8. The apparatus of claim 7, wherein, The first determining module comprises: The determining submodule is configured to determine the candidate user to comprise at least one of: a user whose answering speed is less than or equal to the first threshold value; a user whose answering speed is greater than or equal to the second threshold value; a user whose answering accuracy is greater than or equal to the third threshold value; and a user whose reporting amount is greater than or equal to the fourth threshold value.

9. The apparatus of claim 8, wherein, The recognition unit comprises: The third obtaining module is configured to obtain at least one of a first answering speed and a first answering accuracy of the candidate user interacting with the virtual intelligent user according to the interaction data. The recognition module is configured to determine a category of the candidate user according to at least one of the average answering speed, the average answering accuracy, the first answering speed, and the first answering accuracy.

10. The apparatus of claim 9, wherein, The recognition module comprises: The first recognition submodule is configured to, when the average answering speed is less than or equal to the first threshold value, determine the candidate user as the user having the preset behavior if the first answering speed is less than or equal to the average answering speed. The second recognition submodule is configured to, when the average answering speed is greater than or equal to the second threshold value, determine the candidate user as the user having the preset behavior if the first answering speed is greater than or equal to the average answering speed. The third recognition submodule is configured to, when the average answering accuracy is greater than or equal to the third threshold value, determine the candidate user as the user having the preset behavior if the first answering accuracy is greater than or equal to the average answering accuracy and the first answering speed is greater than or equal to the average answering speed. The fourth recognition submodule is configured to determine the candidate user as the user not having the preset behavior.

11. An electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-5. The computer instructions are used to enable the computer to perform the method of any one of claims 1-5.

12. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, 13. A computer program product comprising a computer program which, when executed by a processor, implements the steps of the method of any one of claims 1-5. ​

Images