Method, device, electronic device and storage medium for detecting cheating behavior

By obtaining account behavior data and using cheat detection models to identify behavior similarity between accounts, the accuracy and efficiency of account cheat detection in the existing technology is solved, and more efficient account management is achieved.

CN111915381BInactive Publication Date: 2024-11-22BEIJING DIDI INFINITY TECH & DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010962524.4
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-09-14
Publication Date
2024-11-22
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The prior art is difficult to effectively detect and identify account cheating, resulting in increased management costs and low accuracy, especially in cases of multi-account registration and high behavioral similarity.

Method used

By obtaining account behavior data, calculating behavior similarity, and using the cheat detection model formed by training, we can identify the relationship between behavior similarity between accounts and cheating behavior, and realize automated detection.

Benefits of technology

It improves the accuracy and efficiency of account cheat detection, reduces manual operation requirements, reduces management costs, and can more effectively identify potential cheat scenarios.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN111915381B_ABST
    Figure CN111915381B_ABST
Patent Text Reader

Abstract

The present disclosure relates to a method, device, electronic device and storage medium for detecting cheating behavior. In one method, first behavior data and second behavior data of a first account and a second account are obtained respectively, and the first behavior data and the second behavior data respectively represent the behavior data of the first account and the second account providing services to a first set and a second set of service objects. Based on the first behavior data and the second behavior data, the behavior similarity between the first account and the second account is determined respectively. Based on the behavior similarity and a cheating detection model formed by training, the cheating behavior of the first account and the second account is detected, and the cheating detection model describes the association between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behavior of multiple accounts. Further, corresponding devices, electronic devices and storage media are provided. In this way, it is possible to detect whether each account has cheating behavior based on the behavior data of each account.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] Various implementations of the present disclosure relate to account management, and more specifically, to a method, device, electronic device, and storage medium for detecting cheating behavior of an account. Background Art

[0002] With the development of computer technology and network technology, corresponding applications have been developed for various types of workers. For example, bicycle operators can use operation and maintenance applications to manage the operation and maintenance services provided to bicycles, couriers can use courier applications to manage package delivery, and so on. Application providers can provide remuneration and / or rewards based on the behavior data of workers. However, some bad users may register multiple accounts. These users may log in to multiple accounts every day and provide services through multiple accounts in order to defraud remuneration and / or rewards. Identifying cheating behavior involves a lot of manual operations and has low accuracy. At this time, how to detect cheating behavior in a more effective way has become a research hotspot. Summary of the invention

[0003] It is expected that a technical solution can be developed and implemented to detect cheating behavior in a more effective manner. It is expected that the technical solution can be compatible with existing applications, detect cheating behavior of application accounts in a more effective manner, and thus manage applications in a more effective manner.

[0004] According to a first aspect of the present disclosure, a method for detecting cheating behavior is provided. In the method, first behavior data and second behavior data of a first account and a second account are obtained respectively, and the first behavior data and the second behavior data respectively represent behavior data of services provided by the first account and the second account to a first set and a second set of service objects. Based on the first behavior data and the second behavior data, respectively, the behavior similarity between the first account and the second account is determined. Based on the behavior similarity and a cheating detection model formed by training, the cheating behavior of the first account and the second account is detected, and the cheating detection model describes the association between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behavior of multiple accounts.

[0005] According to the second aspect of the present disclosure, a device for detecting cheating behavior is provided. The device includes: an acquisition module configured to respectively acquire first behavior data and second behavior data of a first account and a second account, the first behavior data and the second behavior data respectively representing behavior data of providing services to a first set and a second set of service objects by the first account and the second account; a determination module configured to determine the behavior similarity between the first account and the second account based on the first behavior data and the second behavior data respectively; and a detection module configured to detect cheating behavior of the first account and the second account based on the behavior similarity and a cheating detection model formed by training, the cheating detection model describing the association between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behavior of multiple accounts.

[0006] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to the first aspect of the present disclosure.

[0007] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, on which one or more computer instructions are stored, wherein the one or more computer instructions are executed by a processor to implement the method according to the first aspect of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] The features, advantages and other aspects of various implementations of the present disclosure will become more apparent with reference to the following detailed description in conjunction with the accompanying drawings, which illustrate several implementations of the present disclosure in an exemplary and non-limiting manner. In the accompanying drawings:

[0009] Figure 1 A block diagram schematically illustrates an application environment in which an exemplary implementation according to the present disclosure may be used;

[0010] Figure 2 A block diagram schematically shows a process for detecting cheating behavior according to an exemplary implementation of the present disclosure;

[0011] Figure 3 A flowchart of a method for detecting cheating behavior according to an exemplary implementation of the present disclosure is schematically shown;

[0012] Figure 4A , Figure 4B and Figure 4C Schematically showing a block diagram of behavior data according to an exemplary implementation of the present disclosure;

[0013] Figure 5 A block diagram schematically illustrates a data structure of behavior similarity according to an exemplary implementation of the present disclosure;

[0014] Figure 6 A block diagram schematically illustrates a process for determining an object repetition rate according to an exemplary implementation of the present disclosure;

[0015] Figure 7 A block diagram schematically illustrates a process for determining a trajectory repetition rate according to an exemplary implementation of the present disclosure;

[0016] Figure 8 A block diagram schematically illustrates a process for training a cheating detection model according to an exemplary implementation of the present disclosure;

[0017] Fig. 9 A block diagram of a cheating detection model implemented based on a decision tree according to an exemplary implementation of the present disclosure is schematically shown;

[0018] Fig.10 A block diagram schematically shows a cheating detection model based on a neural network according to an exemplary implementation of the present disclosure; and

[0019] Fig.11 A block diagram of a device for detecting cheating behavior according to an exemplary implementation of the present disclosure is schematically shown. DETAILED DESCRIPTION

[0020] The preferred implementations of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although the preferred implementations of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the implementations set forth herein. On the contrary, these implementations are provided to make the present disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0021] As used herein, the term "including" and its variations mean open inclusion, i.e., "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example implementation" and "an implementation" mean "at least one example implementation". The term "another implementation" means "at least one additional implementation". The terms "first", "second", etc. may refer to different or the same objects. Other explicit and implicit definitions may also be included below.

[0022] At present, corresponding applications have been developed for various types of workers, and the providers of the applications can provide remuneration and / or rewards based on the service data of the workers. However, some users may register multiple accounts and log in to multiple accounts to defraud remuneration and / or rewards. For the sake of ease of description, in the context of this disclosure, only the operation and maintenance application of assisting bicycle operation and maintenance will be described as a specific application environment.

[0023] See first Figure 1 The application environment of the exemplary implementation of the present disclosure is described. Figure 1 A block diagram 100 schematically illustrates an application environment in which an exemplary implementation according to the present disclosure may be used. Figure 1 As shown, the staff 110 can be an operation and maintenance staff for maintaining the service object 112 (e.g., bicycle). Specifically, the staff 110 can deliver bicycles to a designated destination according to the needs of the user, disperse the area with excess bicycles to the surrounding area, repair faulty bicycles, and so on.

[0024] Under normal circumstances, each staff member registers an account. The staff member 110 logs in to his account 114, provides services to multiple service objects 112, and receives remuneration 116. When the workload of the staff member 110 meets the minimum requirement, the staff member 110 can be paid a basic salary. When the workload of the staff member 110 reaches a certain level, an additional reward will be paid. However, some bad users may register multiple accounts and make multiple accounts meet the minimum workload requirement to defraud remuneration.

[0025] For example, the worker 110 may register another account 124 and register the workload of providing services to the service object 122, etc., to the account 124. At this time, when the workload of both accounts meets the minimum requirements, the worker 110 will receive two rewards 116 and 126. The above cheating behavior may cause one person to register a large number of users, which increases the management burden of the operation and maintenance application. At this time, the provider of the operation and maintenance application has to spend a lot of human resources and computing resources to detect cheating behavior in order to reduce its own economic losses.

[0026] It will be understood that the bicycle described herein is merely an example of a service object in the context of bicycle operation and maintenance. In the context of the present disclosure, a service object may be any type of serviced object associated with an account. For example, in a courier application, a service object may be a package; in a meal ordering application, a service object may be take-out food; in a security application, a service object may be a check-in device on a patrol route; in a fitness application, a service object may be a clock-in device at a gym, and so on.

[0027] At present, technical solutions for determining cheating behavior have been proposed. For example, it is possible to monitor whether cheating behavior exists based on the motion trajectory of the account. For another example, it is possible to determine whether cheating behavior exists based on a threshold condition. When the behavioral similarity of two accounts is higher than a predetermined threshold condition, it is determined that cheating behavior exists. However, the above technical solutions may cause misjudgment, resulting in the need to distinguish cheating behavior based on a large number of manual operations in the later stage. On the other hand, cheating methods are varied and difficult to identify, and technical solutions based on motion trajectories and / or threshold conditions are difficult to provide accurate detection.

[0028] In order to at least partially solve the deficiencies in the above technical solutions, according to the exemplary implementation of the present disclosure, a technical solution for detecting cheating behavior is proposed. Figure 2 An overview of an exemplary implementation according to the present disclosure is described. Figure 2 A block diagram 200 schematically shows a process for detecting cheating behavior according to an exemplary implementation of the present disclosure. Figure 2 As shown, the behavior of multiple accounts can be monitored. Assume that a first staff member 210 logs in to a first account 214 and provides operation and maintenance services to a first set 212 of service objects. A second staff member 220 logs in to a second account 224 and provides operation and maintenance services to a second set 222 of service objects. At this time, the first behavior data 216 and the second behavior data 226 of the first account 214 and the second account 224 can be obtained respectively.

[0029] A behavior similarity 230 between the first behavior data 216 and the second behavior data 226 may be determined. Further, based on the behavior similarity 230 and the cheating detection model 240, it may be determined whether the accounts of the two staff members have cheating behavior 250. In other words, it may be determined based on the predefined cheating detection model 240 whether the first staff member 210 and the second staff member 220 are the same staff member.

[0030] Here, the behavior data may include various information about the operation and maintenance services provided by the staff, and may fully reflect the operation and maintenance service process. By determining the behavior similarity 230 based on the behavior data, the similarity between the behaviors of two staff members may be accurately determined, and then whether the two staff members are the same staff members may be determined. In this way, by performing various analyses on massive amounts of behavior data, potential cheating scenarios may be mined, and cheating behaviors may be effectively identified and operation and maintenance costs may be reduced.

[0031] In the following, we will refer to Figure 3 More details of exemplary implementations of the present disclosure are described. Figure 3A flowchart of a method 300 for detecting cheating behavior according to an exemplary implementation of the present disclosure is schematically shown. At box 310, first behavior data 216 and second behavior data 226 of a first account 214 and a second account 224 are obtained respectively. It will be understood that the first behavior data 216 and the second behavior data 226 here represent the behavior data of providing services to a first set 212 and a second set 222 of service objects by the first account 214 and the second account 224 respectively. Assuming that the first staff member 210 registers the workload of repairing a bicycle to the first account 214, the first behavior data includes data on the related behavior of repairing the bicycle.

[0032] In the context of the present disclosure, behavior data may include many aspects, for example, service records, check-in records, heartbeat records, etc. FIG. 4A to FIG. 4C Describing the content of the behavior data It will be understood that the format of the behavior data is the same for all staff members, and hereinafter, only the first behavior data 216 of the first staff member 210 will be described as an example. Figure 4A A block diagram 400A schematically shows a data structure of a service record according to an exemplary implementation of the present disclosure. Figure 4A As shown, service record 410A may include: account identifier 412A, object identifier 414A, service type 416A, city 418A, location 420A, and update time 422A.

[0033] For the first staff member 210, the account identifier 412A may represent the identifier of the first staff member 210 (e.g., work number); the object identifier 414A may represent the identifier of the service object provided by the first staff member 210 (e.g., bicycle number); the service type 416A may represent the type of service provided by the first staff member 210 (e.g., moving or repair, etc.); the city 418A may represent the city where the first staff member 210 is located when providing the service; the location 420A may represent the location where the first staff member 210 is located when providing the service; and the update time 422A may represent the time when this service record is updated.

[0034] According to an exemplary implementation of the present disclosure, a service record may be added each time the first staff member 210 serves a new service object. At this time, the behavior data will include a set of identifiers of all service objects processed by the staff member. In other words, the first behavior data 216 and the second behavior data 226 include the first set 212 and the second set 222, respectively.

[0035] will understand, Figure 4AThe service record 410A shown in FIG. 4 is merely illustrative. According to an exemplary implementation of the present disclosure, the service record 410A may include more or fewer fields. For example, one service record may not include the city 418A, and another service record may include an identifier of the work group to which the staff member belongs. Using the exemplary implementation of the present disclosure, various aspects of information about each service operation can be fully recorded, thereby facilitating the acquisition of richer behavioral data.

[0036] According to an exemplary implementation of the present disclosure, the behavior data may include a check-in record. Figure 4B A block diagram 400B schematically shows a data structure of a check-in record according to an exemplary implementation of the present disclosure. Figure 4B As shown, the check-in record 410B may include: an identifier 412B of the first account, a check-in time 414B, a check-out time 416B, a check-in city 418B, and a record update time 420B. According to an exemplary implementation of the present disclosure, the check-in data and check-out data of the staff members when they go to and from get off work every day can be obtained, which helps to extract service records related to labor during work.

[0037] will understand, Figure 4B The check-in record 410B shown in is merely illustrative. According to an exemplary implementation of the present disclosure, the check-in record 410B may include more or fewer fields. For example, one check-in record may not include the city 418B, and another service record may include an identifier of the operation and maintenance enterprise to which the staff member belongs. Using the exemplary implementation of the present disclosure, the commuting information of the staff member can be obtained, which is helpful for subsequent filtering of the behavior data.

[0038] According to an exemplary implementation of the present disclosure, the behavior data may include heartbeat records. Figure 4C A block diagram 400C schematically shows a data structure of a heartbeat record according to an exemplary implementation of the present disclosure. Figure 4C As shown, the heartbeat record 410C may include: an identifier 412C of the first account, a city 414C associated with the first account, a real-time location 416C (eg, GPS data) associated with the first account, and a data update time 418C.

[0039] will understand, Figure 4CThe heartbeat record 410C shown in is merely illustrative. According to an exemplary implementation of the present disclosure, the service record 410C may include more or fewer fields. For example, one service record may not include the city 418A, and another service record may include an identifier of the work group to which the staff member belongs. According to an exemplary implementation of the present disclosure, heartbeat records may be collected at predetermined time intervals (e.g., every 10 seconds or other time intervals). Using the exemplary implementation of the present disclosure, real-time location information of the staff member may be obtained, which in turn helps to determine cheating behavior based on the various location information in the movement trajectory of the staff member.

[0040] It will be understood that there may be noise data in the collected raw data, and therefore the collected raw behavior data should be filtered. Using the exemplary implementation of the present disclosure, the noise in the behavior data can be eliminated by filtering operations. In this way, it can be ensured that the basis for determining the behavior similarity is accurate data that can truly reflect the working status of the staff. According to the exemplary implementation of the present disclosure, the behavior data can be filtered based on the check-in time and the check-out time.

[0041] According to an exemplary implementation of the present disclosure, duplicate data in behavioral data can be removed. For example, based on the nature of the work, a large amount of duplicate data may appear in the behavioral data of warehouse staff. For example, staff can be divided into: porters, who are responsible for transporting faulty vehicles to the warehouse; maintenance personnel, who are responsible for repairing faulty vehicles; and acceptance personnel, who are responsible for checking whether the vehicle fault has been eliminated. At this time, a large amount of duplicate information will appear in the service object, location and other fields of these three types of staff. Given that these behavioral data are normal data and do not involve cheating, such duplicate data should be removed to avoid misjudgment.

[0042] According to an exemplary implementation of the present disclosure, null value data in the behavior data can be removed. During the service process of the staff, null value data may appear for certain reasons. For example, the QR code on the bicycle may be unclear, resulting in the inability to identify the object identifier, the positioning module may temporarily fail due to occlusion and other reasons, and so on, resulting in null value fields in the behavior data. At this time, the records including null value fields should be cleared. According to an exemplary implementation of the present disclosure, empty abnormal data in the behavior data can be removed. For example, the positioning module may have a large drift and cause position abnormalities, and so on. At this time, the abnormalities in the behavior data should be detected and the records including the abnormal data should be cleared.

[0043] Already seen Figure 4A , Figure 4B and Figure 4C Describes the details of the behavior data. In the following, we will return Figure 3Describes how to determine behavioral similarity based on behavioral data. Figure 3 At block 320, based on the first behavior data 216 and the second behavior data 226, the behavior similarity 230 between the first account 214 and the second account 224 is determined. According to an exemplary implementation of the present disclosure, the similarity may include multiple aspects, which will be described below. Figure 5 Provide more details.

[0044] Figure 5 A block diagram 500 schematically shows a data structure of behavior similarity according to an exemplary implementation of the present disclosure. Figure 5 As shown, the behavior similarity 230 may include one or more of the following aspects: object repetition rate 512, object repetition time 514, track repetition rate 516, and track repetition time 518. Figure 6 Further details regarding determining the object repetition rate 512 are described.

[0045] Figure 6 A block diagram 600 of a process for determining an object repetition rate according to an exemplary implementation of the present disclosure is schematically shown. The first set 212 and the second set 222 may be determined from the behavior data, respectively. For example, Figure 4A The service record 410A shown in FIG. 4 shows that the first set 212 includes 10 service objects: V1, V2, V3, V4, V5, V6, V7, V8, V9, and V10. Similarly, the second set 222 includes 11 service objects: V1, V3, V5, V7, V9, V12, V13, V14, V15, V16, and V17.

[0046] It can be determined that the object intersection 610 between the first set 212 and the second set 222 includes 5 elements: V1, V3, V5, V7, and V9. The object repetition rate 512 can be determined based on the object intersection 610. For example, the object repetition rate 512 can be determined based on the ratio of the number of service object elements included in the object intersection 610 to the first number of service objects included in the first set 212. At this time, the object repetition rate = 5 / 10 = 0.5.

[0047] It will be understood that the formula for determining the object repetition rate 512 above is merely illustrative. According to an exemplary implementation of the present disclosure, the object repetition rate 512 may be determined based on the ratio of the number of service objects included in the object intersection 610 to the second number of service objects included in the second set 222 (or the minimum or maximum value of the first number and the second number).

[0048] It will be understood that the two accounts may involve the same service objects. However, if two accounts handle a large number of the same service objects within one day, there is reason to suspect that the two accounts are suspected of cheating. For example, for staff members who carry bicycles, there may be a situation where a staff member logs in to two accounts, uses one account to carry the bicycle from A to B, and then uses another account to carry the bicycle from B back to A (or other locations). When rewards are paid according to the carrying distance and the number of bicycles carried, the cheating behavior of the staff member can obtain a large amount of rewards. However, the above-mentioned carrying behavior does not achieve the purpose of delivering bicycles according to user needs, and therefore should be prohibited.

[0049] By using the exemplary implementation of the present disclosure, the degree of duplication of multiple service objects processed by two accounts can be determined in a quantitative manner. In this way, the similarity of the behavior of each account can be described in a more comprehensive manner, thereby facilitating the detection of the cheating described above.

[0050] According to an exemplary implementation of the present disclosure, the behavior similarity 230 further includes an object repetition time 514. A predetermined threshold value (e.g., 0.3 or other numerical values) may be set. If the object repetition rate of the behavior data of the two accounts within a predetermined time period is higher than the threshold value, the object repetition time may be increased. Assuming that the predetermined time period is 1 day, the behavior data of the two accounts within multiple days (e.g., a week or other days) may be counted. If it is found that the object repetition rate of the two accounts within one day is higher than the predetermined threshold, the object repetition time may be increased by one. If the object repetition rate within 5 days of a week is higher than the predetermined threshold, the object repetition time may be set to 5 at this time. According to an exemplary implementation of the present disclosure, the object repetition time may be normalized, and the object repetition time may be set to 5 / 7 at this time.

[0051] By using the exemplary implementation of the present disclosure, the long-term repetition degree of the behavior of two accounts can be determined in a quantitative manner. In this way, the similarity of the behavior of each account can be described in a more comprehensive manner.

[0052] According to an exemplary implementation of the present disclosure, the behavior similarity 230 may include a trajectory repetition rate 516. Figure 7 Describes more details about determining the trajectory repetition rate. Figure 7 A block diagram 700 schematically illustrates a process for determining a trajectory repetition rate according to an exemplary implementation of the present disclosure. Figure 7 As shown, a first track 710 of the first account 214 and a second track 720 of the second account 224 are determined based on the first behavior data 216 and the second behavior data 226, respectively.

[0053] For example, one can Figure 4A In the service record 410A shown, it is determined that the first track 710 includes 10 positions: P1, P2, P3, P4, P5, P6, P7, P8, P9, and P10. Similarly, it can be determined that the second track 720 includes 9 positions: P1, P3, P4, P5, P6, P12, P13, P14, and P17. It can be determined that the common sub-track 730 between the first track 710 and the second track 720 includes 5 elements: P1, P3, P4, P5, and P6. The track repetition rate 516 can be determined based on the common sub-track 730. For example, the track repetition rate 516 can be determined based on the ratio of the number of positions included in the common sub-track 730 to the first number of positions included in the first track 710. At this time, the object repetition rate = 4 / 10 = 0.4.

[0054] It will be understood that the formula for determining the trajectory repetition rate 516 in the above text is merely illustrative. According to an exemplary implementation of the present disclosure, the trajectory repetition rate 516 may be determined based on the ratio of the number of positions included in the longest common sub-trajectory 730 to the second number of positions included in the second trajectory 720 (the minimum value or the maximum value of the first number and the second number). According to an exemplary implementation of the present disclosure, when there are multiple common subsequences between two trajectories, the trajectory repetition rate may be determined based on the longest common subsequence, and the trajectory repetition rate may also be determined based on the number of common subsequences or the total number of common positions included in each common subsequence.

[0055] It will be understood that the accounts of two staff members in the same work group may involve similar trajectories. However, if most of the trajectories of two work accounts within a day are exactly the same, there is reason to suspect that the two accounts are suspected of cheating. For example, for a staff member who carries a bicycle, a staff member may log in to two accounts and use the two accounts to carry the bicycles so that both accounts reach the minimum workload, thereby defrauding compensation. Using the exemplary implementation of the present disclosure, the degree of repetition of the motion trajectories of two accounts can be determined in a quantitative manner. In this way, the similarity of the behavior of each account can be described in a more comprehensive manner, which helps to detect the cheating described above.

[0056] According to an exemplary implementation of the present disclosure, the behavior similarity 230 may further include a trajectory repetition time 518. If it is determined that the trajectory repetition rate associated with the predetermined time period is higher than the predetermined trajectory repetition rate threshold, the trajectory repetition time is increased. The trajectory repetition time 518 may be determined based on a method similar to the determination of the object repetition time described above. For example, the predetermined threshold may be set to 0.7 (or other values), and if the trajectory repetition rate within 3 days of the week is higher than the predetermined threshold, then the trajectory repetition time 518 may be set to 3 at this time. According to an exemplary implementation of the present disclosure, the trajectory repetition time 518 may be normalized, and the trajectory repetition time 518 may be set to 3 / 7 at this time.

[0057] By using the exemplary implementation of the present disclosure, the long-term repetition degree of the movement trajectories of two accounts can be determined in a quantitative manner. In this way, the similarity of the behavior of each account can be described in a more comprehensive manner.

[0058] See above Figures 5 to 7 The process of determining the behavior similarity 230 is described below. Figure 3 Describe how cheating is identified. Figure 3 At block 330, based on the behavior similarity 230 and the cheating detection model 240 formed by training, the cheating behavior of the first account 214 and the second account 224 is detected. Here, the cheating detection model 240 describes the association between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behavior of the multiple accounts.

[0059] According to an exemplary implementation of the present disclosure, a cheating detection model 240 may be obtained based on a training process. Data related to multiple accounts in a training set may be used for training. Specifically, multiple reference accounts may be pre-specified in the training set, for example, multiple accounts actually registered in an operation and maintenance application may be used as reference accounts. Multiple training samples may be obtained based on multiple reference accounts. Figure 8 Describe in more detail. Figure 8 A block diagram 800 of a process for training a cheating detection model according to an exemplary implementation of the present disclosure is schematically shown. For two reference accounts among a plurality of reference accounts, a training sample (e.g., Figure 8 The two behavior data may be determined respectively according to the method described above, and the behavior similarity 812 between the two behavior data may be determined.

[0060] Furthermore, it is possible to determine whether the two accounts are accounts of two real staff members through the work records of actual operation and maintenance data, the actual evaluations of the staff members' colleagues, and the leaders of the staff members at all levels. If it is determined that the two accounts belong to two real staff members, the cheating behavior 814 can be set to "false"; if it is determined that the two accounts belong to the same staff member, the cheating behavior 814 can be set to "true". At this time, the training sample 810 will include the behavior similarity 812 (represented by a multidimensional vector) and the cheating behavior 814 (represented by a Boolean value).

[0061] For any two reference accounts among the multiple reference accounts, training samples can be obtained in a similar manner. At this time, multiple training samples can be obtained, and each training sample will include behavior similarity and cheating behavior. Further, the cheating detection model 240 can be trained using multiple training samples.

[0062] According to an exemplary implementation of the present disclosure, a cheating detection model 240 can be obtained based on a variety of methods currently known and / or to be developed in the future. After the training process is completed, when the behavioral similarity between two reference persons in multiple reference accounts is input to the trained cheating detection model, the cheating behavior of the two reference persons can be output. That is, the output result of the trained cheating detection model is consistent with the cheating behavior in the initial training sample. According to an exemplary implementation of the present disclosure, multiple training samples can be divided into a training set and a test set. The cheating detection model can be trained using samples in the training set, and the trained cheating detection model can be tested to see if it is accurate using samples in the test set.

[0063] It will be understood that although the data related to the reference account in the training set is used to perform training, the training set can cover a large number of accounts and have diversity. At this time, the cheating detection model 240 obtained by training can reflect the correlation between the behavior similarity of providing services by multiple accounts and the cheating behavior of multiple accounts.

[0064] According to the exemplary implementation of the present disclosure, the cheating detection model 240 can be obtained based on a variety of machine learning techniques, for example, the cheating detection model 240 can be obtained based on a decision tree or a neural network. It will be understood that the essence of cheating detection is to classify multiple behavior similarities into two categories of "cheating" and "non-cheating". The decision tree is a commonly used classification scheme that can divide the entire space of input data into a tree structure, and each leaf node represents a classification.

[0065] Fig. 9 A block diagram 900 of a cheating detection model based on a decision tree according to an exemplary implementation of the present disclosure is schematically shown. Fig. 9As shown, node 910 represents the root node of the decision tree, and there may be one or more levels below the root node, where non-leaf nodes represent an output, and each leaf node represents a final classification. For example, the lower layer of non-leaf node 920 includes leaf nodes 930 and 932, which respectively represent that the input behavior similarity is classified as "cheating" or "cheating". The root node 910 may include another child node 922, and the lower layer nodes of the child node 922 are omitted for simplicity.

[0066] exist Fig. 9 In each node, "Number of samples" indicates the number of samples involved in this level. For example, the number of samples of the root node 910 is 64, indicating that 64 samples are used to generate the decision tree. The number of samples of the non-leaf node 920 is 23, indicating that 23 samples are output to this branch. The number of samples of the leaf node 930 is 22, indicating that 22 samples are classified as "non-cheating"; the number of samples of the leaf node 932 is 1, indicating that 1 sample is classified as "cheating". In each node, "Value" indicates the number of samples classified into the lower-level child nodes. For example, the value of the root node 910 is "[41, 23]", indicating that 23 samples are output to the left branch and 41 samples are output to the right branch.

[0067] It will be understood that the above only refers to Fig. 9 The general principles of decision trees are described. For simplicity, further details about creating decision trees will be omitted in the context of the present disclosure. Decision trees can be created based on more techniques related to decision trees that are currently known and / or will be developed in the future. Using the exemplary implementation of the present disclosure, a decision tree can be established based on the behavioral similarities and related cheating behaviors from multiple reference accounts. In this way, the decision tree can fully retain the historical experience of cheating behaviors and serve as a basis for predicting whether other accounts are suspected of cheating.

[0068] According to an exemplary implementation of the present disclosure, the cheating detection model 240 may be determined based on a neural network. Fig.10 A block diagram 1000 of a cheating detection model based on a neural network according to an exemplary implementation of the present disclosure is schematically shown. According to an exemplary implementation of the present disclosure, a cheating detection function 1010 (including multiple influencing factors 1020, 1022, ..., and 1024) can be set, and corresponding weights W1, W2, ..., and Wm can be set for each influencing factor. The cheating detection model 240 can be iteratively trained using multiple training samples. Specifically, Figure 8 Each component of the behavior similarity 812 shown is input to Fig.10The object repetition rate 512, object repetition time 514, trajectory repetition rate 516 and trajectory repetition time 518 in the embodiment of the present invention can be input into the cheating behavior 814. Fig.10 The cheating behavior 1030 in the example is detected to obtain the specific value of each weight. In this way, a cheating detection model based on a neural network can be obtained.

[0069] By using the exemplary implementation of the present disclosure, a neural network can be established based on the behavior similarities and related cheating behaviors from multiple reference accounts. In this way, each coefficient in the neural network can fully retain the historical experience of cheating behaviors and serve as a basis for predicting whether other accounts are suspected of cheating.

[0070] In the case where the cheating detection model 240 has been obtained, it can be determined whether the two accounts are suspected of cheating based on the behavior similarity 230 between the first account 214 and the second account 224 that currently need to be detected. Specifically, the behavior similarity between the two accounts to be judged can be input into the cheating detection model 240 to obtain the cheating behavior. If the cheating detection model 240 outputs "cheating", it is determined that the two accounts are logged in by the same user. Using the exemplary implementation of the present disclosure, it can be determined whether there is cheating behavior based on the successful historical experience in the cheating detection model 240. In this way, the manual labor involved in the detection process can be reduced, and accounts suspected of cheating behavior can be detected more accurately.

[0071] According to the exemplary implementation of the present disclosure, if cheating behavior is detected, the first account and the second account can be identified as cheating accounts. Specifically, the administrator of the operation and maintenance application can be notified to suspend the use of the cheating account, and the leaders of the staff using the cheating account can be notified to review so as to take corresponding measures. Using the exemplary implementation of the present disclosure, accounts suspected of cheating behavior can be marked in time to prevent these accounts from disrupting the normal operation and maintenance order.

[0072] It will be understood that although the exemplary implementation according to the present disclosure is described above in an application environment for operation and maintenance of shared bicycles, the method 300 described above may also be suitable for other application environments. For example, in a meal ordering application, multiple customers in an office building order meals at the same restaurant, and multiple takeaway tasks will be generated. Some delivery personnel may log in to multiple meal delivery applications at the same time, perform meal delivery tasks and defraud remuneration. For another example, in a security application, a security company requires security personnel to patrol regularly and set up sign-in equipment at designated key locations. Some security personnel may log in to other people's accounts, sign in on their behalf and defraud remuneration. For another example, in a fitness application, a coach may ask students to punch in at a designated gym. Some students may log in to other people's accounts and punch in on their behalf. The method 300 described above can be used to detect the above-mentioned cheating.

[0073] See above Figures 2 to 10 The process of the method for detecting cheating behavior is described. According to an exemplary implementation of the present disclosure, a device for detecting cheating behavior is provided. The device includes: an acquisition module, configured to respectively acquire first behavior data and second behavior data of a first account and a second account, the first behavior data and the second behavior data respectively representing behavior data of providing services by the first account and the second account to a first set and a second set of service objects; a determination module, configured to determine the behavior similarity between the first account and the second account based on the first behavior data and the second behavior data, respectively; and a detection module, configured to detect cheating behavior of the first account and the second account based on the behavior similarity and a cheating detection model formed by training, the cheating detection model describing the association between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behavior of multiple accounts.

[0074] According to an exemplary implementation of the present disclosure, the first behavior data and the second behavior data respectively include a first set and a second set of service objects, the behavior similarity includes an object repetition rate, and a determination module configured to include: an intersection determination module configured to determine the object intersection between the first set and the second set; and an object repetition rate determination module configured to determine the object repetition rate based on the object intersection.

[0075] According to an exemplary implementation of the present disclosure, the behavior similarity further includes object repetition time, and the determination module includes: a repetition time determination module, configured to increase the object repetition time in response to determining that the object repetition rate associated with the predetermined time period is higher than a predetermined object repetition rate threshold.

[0076] According to an exemplary implementation of the present disclosure, the behavior similarity includes a trajectory repetition rate, and the determination module includes: a trajectory determination module, configured to determine a first trajectory of a first account and a second trajectory of a second account based on first behavior data and second behavior data, respectively; a common subsequence determination module, configured to determine a common sub-trajectory between the first trajectory and the second trajectory; and a trajectory repetition rate determination module, configured to determine the trajectory repetition rate based on the common sub-trajectory.

[0077] According to an exemplary implementation of the present disclosure, the behavior similarity further includes a trajectory repetition time, and the determination module includes: a trajectory repetition time determination module configured to increase the trajectory repetition time in response to determining that a trajectory repetition rate associated with a predetermined time period is higher than a predetermined trajectory repetition rate threshold.

[0078] According to an exemplary implementation of the present disclosure, the acquisition module includes a service record acquisition module configured to acquire: an identifier of the first account, an identifier of a service object provided by the first account, and a record update time.

[0079] According to an exemplary implementation of the present disclosure, the record acquisition module is further configured to acquire: the type of service provided by the first account, the city where the service is provided by the first account, and the location associated with the first account.

[0080] According to an exemplary implementation of the present disclosure, the acquisition module includes a sign-in record acquisition module configured to acquire: an identifier of the first account, a sign-in time, a sign-out time, and a record update time.

[0081] According to an exemplary implementation of the present disclosure, the check-in record acquisition module is further configured to acquire the check-in city.

[0082] According to an exemplary implementation of the present disclosure, the acquisition module includes a heartbeat record acquisition module configured to acquire: an identifier of the first account, a location associated with the first account, and a data update time.

[0083] According to an exemplary implementation of the present disclosure, the heartbeat record acquisition module is further configured to acquire a city associated with the first account.

[0084] According to an exemplary implementation of the present disclosure, the device further includes a filtering module configured to filter the first behavior data based on at least any one of the following before determining the behavior similarity: removing duplicate data, removing null value data, and removing abnormal data.

[0085] According to an exemplary implementation of the present disclosure, the acquisition module further includes: a similarity determination module, configured to determine the behavioral similarity of multiple reference accounts in providing services to their respective reference service object sets; a cheating behavior determination module, configured to determine the cheating behavior of multiple reference accounts; and a training module, configured to train a cheating detection model based on the behavioral similarity and the cheating behavior, so that the trained cheating detection model outputs the cheating behavior of two reference persons in the multiple reference accounts based on the behavioral similarity between the two reference persons.

[0086] According to an exemplary implementation of the present disclosure, the cheating detection model includes at least any one of the following: a classification regression tree model and a neural network model.

[0087] According to an exemplary implementation of the present disclosure, the detection module further includes: an input module, configured to input behavior similarity into a cheating detection model to obtain cheating behavior; and a login cheating behavior determination module, configured to determine that the first account and the second account are logged in by the same user in response to the cheating behavior indicating "cheating".

[0088] According to an exemplary implementation of the present disclosure, the apparatus further includes an identification module configured to identify the first account and the second account as cheating accounts.

[0089] According to an exemplary implementation of the present disclosure, an electronic device is provided, including: a memory and a processor; wherein the memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method described above.

[0090] Fig.11 1 shows a block diagram of a computing device / server 1100 in which one or more embodiments of the present disclosure may be implemented. It should be understood that Fig.11 The computing device / server 1100 shown is merely exemplary and should not constitute any limitation on the functionality and scope of the embodiments described herein.

[0091] like Fig.11 As shown, computing device / server 1100 is in the form of a general-purpose computing device. Components of computing device / server 1100 may include, but are not limited to, one or more processors or processing units 1110, memory 1120, storage device 1130, one or more communication units 1140, one or more input devices 1150, and one or more output devices 1160. Processing unit 1110 may be an actual or virtual processor and is capable of performing various processes according to a program stored in memory 1120. In a multi-processor system, multiple processing units execute computer executable instructions in parallel to increase the parallel processing capabilities of computing device / server 1100.

[0092] The computing device / server 1100 typically includes a plurality of computer storage media. Such media may be any available media accessible to the computing device / server 1100, including but not limited to volatile and non-volatile media, removable and non-removable media. The memory 1120 may be a volatile memory (e.g., registers, caches, random access memory (RAM)), a non-volatile memory (e.g., a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), flash memory), or some combination thereof. The storage device 1130 may be a removable or non-removable medium, and may include a machine-readable medium, such as a flash drive, a disk, or any other medium, which may be capable of being used to store information and / or data (e.g., training data for training) and may be accessed within the computing device / server 1100.

[0093] The computing device / server 1100 may further include additional removable / non-removable, volatile / non-volatile storage media. Fig.11 As shown in , a disk drive for reading or writing from a removable, non-volatile disk (e.g., a "floppy disk") and an optical drive for reading or writing from a removable, non-volatile optical disk may be provided. In these cases, each drive may be connected to the bus (not shown) by one or more data media interfaces. Memory 1120 may include a computer program product 1125 having one or more program modules that are configured to perform various methods or actions of various embodiments of the present disclosure.

[0094] The communication unit 1140 enables communication with other computing devices via a communication medium. Additionally, the functionality of the components of the computing device / server 1100 can be implemented in a single computing cluster or multiple computing machines that can communicate via a communication connection. Thus, the computing device / server 1100 can operate in a networked environment using a logical connection with one or more other servers, a network personal computer (PC), or another network node.

[0095] Input device 1150 may be one or more input devices, such as a mouse, keyboard, trackball, etc. Output device 1160 may be one or more output devices, such as a display, speaker, printer, etc. Computing device / server 1100 may also communicate with one or more external devices (not shown) as needed, such as storage devices, display devices, etc., with one or more devices that enable users to interact with computing device / server 1100, or with any device that enables computing device / server 1100 to communicate with one or more other computing devices (e.g., network card, modem, etc.) through communication unit 1140. Such communication may be performed via an input / output (I / O) interface (not shown).

[0096] According to an exemplary implementation of the present disclosure, a computer-readable storage medium is provided, on which one or more computer instructions are stored, wherein the one or more computer instructions are executed by a processor to implement the method described above.

[0097] Various aspects of the present disclosure are described herein with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products implemented according to the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of each box in the flowchart and / or block diagram can be implemented by computer-readable program instructions.

[0098] These computer-readable program instructions can be provided to a processing unit of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processing unit of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0099] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.

[0100] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple implementations of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of a module, program segment or instruction includes one or more executable instructions for realizing the logical function of the specification. In some implementations as replacements, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be realized by a special hardware-based system that performs the function or action of the specification, or can be realized by a combination of special hardware and computer instructions.

[0101] The above descriptions of various implementations of the present disclosure are exemplary, non-exhaustive, and not limited to the disclosed implementations. Many modifications and variations will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described implementations. The selection of terms used herein is intended to best explain the principles of the implementations, practical applications, or improvements to the technology in the marketplace, or to enable other persons of ordinary skill in the art to understand the implementations disclosed herein.

Claims

1. A method for detecting cheating behavior, comprising: Acquire first behavior data and second behavior data of a first account and a second account respectively, wherein the first behavior data and the second behavior data respectively represent behavior data of providing services provided by the first account and the second account to a first set and a second set of service objects; The service object includes a bicycle, and the service provided includes transporting the bicycle; Determining the behavior similarity between the first account and the second account based on the first behavior data and the second behavior data, respectively; the first behavior data and the second behavior data respectively include the first set and the second set of service objects, and the behavior similarity includes object repetition rate, object repetition time, trajectory repetition rate and trajectory repetition time; Determining the object repetition rate includes: determining an object intersection between the first set and the second set, and determining the object repetition rate based on the object intersection; in response to determining that the object repetition rate associated with a predetermined time period is higher than a predetermined object repetition rate threshold, increasing the object repetition time; determining the trajectory repetition rate includes: determining a first trajectory of the first account and a second trajectory of the second account based on the first behavior data and the second behavior data, respectively, determining a common sub-trajectory between the first trajectory and the second trajectory, and determining the trajectory repetition rate based on the common sub-trajectory; in response to determining that the trajectory repetition rate associated with the predetermined time period is higher than a predetermined trajectory repetition rate threshold, increasing the trajectory repetition time; and Based on the behavior similarity and the cheating detection model formed by training, the cheating behaviors of the first account and the second account are detected, wherein the cheating detection model describes the correlation between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behaviors of the multiple accounts; the cheating detection model includes at least one of a classification regression tree model and a neural network model; when training the cheating detection model, corresponding weights are set for the influencing factors of the object repetition rate, the object repetition time, the trajectory repetition rate and the trajectory repetition time.

2. The method according to claim 1, wherein obtaining the first behavior data comprises obtaining a service record of the first account, wherein the service record comprises: The identifier of the first account, the identifier of the service object provided by the first account, and the record update time.

3. The method according to claim 2, wherein the service record further comprises: The type of service provided by the first account, the city provided by the first account, and the location associated with the first account.

4. The method according to claim 1, wherein obtaining the first behavior data comprises obtaining a check-in record of the first account, wherein the check-in record comprises: The identifier of the first account, the check-in time, the check-out time, and the record update time.

5. The method according to claim 4, wherein the check-in record further comprises: Sign in city.

6. The method according to claim 1, wherein obtaining the first behavior data comprises obtaining a heartbeat record of the first account, wherein the heartbeat record comprises: An identifier of the first account, a location associated with the first account, and a data update time.

7. The method according to claim 6, wherein the heartbeat data further comprises: The city associated with the first account. 8 . The method according to claim 1 , further comprising filtering the first behavior data based on at least any one of the following before determining the behavior similarity: removing duplicate data, removing null value data, and removing abnormal data.

9. The method according to claim 1, wherein detecting the cheating behavior of the first account and the second account comprises: Inputting the behavior similarity into the cheating detection model to obtain the cheating behavior; as well as In response to the cheating behavior indicating “cheating”, it is determined that the first account and the second account are logged in by the same user.

10. The method according to claim 9, further comprising: The first account and the second account are identified as cheating accounts.

11. A device for detecting cheating behavior, comprising: An acquisition module configured to acquire first behavior data and second behavior data of a first account and a second account, respectively, wherein the first behavior data and the second behavior data respectively represent behavior data of providing services provided by the first account and the second account to a first set and a second set of service objects; The service object includes a bicycle, and the service provided includes transporting the bicycle; a determination module configured to determine the behavior similarity between the first account and the second account based on the first behavior data and the second behavior data, respectively; the first behavior data and the second behavior data respectively include the first set and the second set of service objects, and the behavior similarity includes an object repetition rate, an object repetition time, a trajectory repetition rate, and a trajectory repetition time; Determining the object repetition rate includes: determining an object intersection between the first set and the second set, and determining the object repetition rate based on the object intersection; in response to determining that the object repetition rate associated with a predetermined time period is higher than a predetermined object repetition rate threshold, increasing the object repetition time; determining the trajectory repetition rate includes: determining a first trajectory of the first account and a second trajectory of the second account based on the first behavior data and the second behavior data, respectively, determining a common sub-trajectory between the first trajectory and the second trajectory, and determining the trajectory repetition rate based on the common sub-trajectory; in response to determining that the trajectory repetition rate associated with the predetermined time period is higher than a predetermined trajectory repetition rate threshold, increasing the trajectory repetition time; and A detection module is configured to detect cheating behaviors of the first account and the second account based on the behavior similarity and a cheating detection model formed by training, wherein the cheating detection model describes the correlation between the behavior similarity of multiple accounts providing services to their respective service object sets and the cheating behaviors of the multiple accounts; the cheating detection model includes at least one of a classification regression tree model and a neural network model; when training the cheating detection model, corresponding weights are set for the influencing factors of the object repetition rate, the object repetition time, the trajectory repetition rate, and the trajectory repetition time.

12. An electronic device comprising: Memory and processor; The memory is used to store one or more computer instructions, wherein the one or more computer instructions are executed by the processor to implement the method according to any one of claims 1 to 10.

13. A computer-readable storage medium having one or more computer instructions stored thereon, wherein the one or more computer instructions are executed by a processor to implement the method according to any one of claims 1 to 10.