Abnormal behavior recognition method and device, nonvolatile storage medium and electronic equipment

Abstract

The invention discloses an abnormal behavior recognition method and device, a nonvolatile storage medium and electronic equipment. The method comprises the following steps: after a consumption behavior executed by a target object on a video playing device is detected, determining object behavior data of the target object in a preset time period; determining a user behavior portrait corresponding to a login user logging in the video playing device, wherein the user behavior portrait comprises distribution conditions of various behaviors implemented through the video playing device and corresponding to the login user; determining a behavior score of the object behavior data according to the user behavior portrait, wherein the behavior score is used for reflecting the similarity between the behavior data and the user behavior portrait; and when the behavior score is not higher than a score threshold corresponding to the user behavior portrait, determining that the consumption behavior is an abnormal behavior. According to the method and the device, the technical problem that benefits of audiences and operators are lost due to the fact that whether the consumption behavior executed by the audiences on the television is the abnormal consumption behavior cannot be determined in related technologies is solved.

Description

Technical Field

[0001] The present application relates to the field of electronic digital data processing, and in particular, to a method and device for identifying abnormal behaviors, a non-volatile storage medium, and an electronic device. Background Art

[0002] In the related art, for the consumption behaviors performed by viewers on a television, it is impossible to determine whether the consumption behavior is an abnormal consumption behavior, such as high-frequency consumption or accidental consumption by minors, etc., which has a negative impact on the interests of both viewers and operators.

[0003] In view of the above problems, no effective solution has been proposed yet. Summary of the Invention

[0004] Embodiments of the present application provide a method and device for identifying abnormal behaviors, a non-volatile storage medium, and an electronic device, so as to at least solve the technical problem that the interests of both viewers and operators are damaged due to the inability to determine whether the consumption behavior performed by a viewer on a television is an abnormal consumption behavior in the related art.

[0005] According to one aspect of the embodiments of the present application, a method for identifying abnormal behaviors is provided, including: after detecting a consumption behavior performed by a target object on a video playback device, determining object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; determining a user behavior portrait corresponding to the logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; determining a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and determining that the consumption behavior is an abnormal behavior when the behavior score is not higher than the score threshold corresponding to the user behavior portrait.

[0006] Optionally, the user behavior portrait includes an operation behavior portrait, a viewing behavior portrait, and a consumption behavior portrait; the method further includes: clustering the behavior data of multiple users to obtain operation behavior types, viewing behavior types, and consumption behavior types, where the behavior data includes operation behavior data, viewing behavior data, and consumption behavior data; obtaining historical user data of the logged-in user, and determining the user behavior portrait based on the historical user data and the operation behavior types, viewing behavior types, and consumption behavior types, where the historical user data includes historical operation behavior data, historical viewing behavior data, and historical consumption behavior data.

[0007] Optionally, determining the behavior score of object behavior data based on the user behavior portrait includes: determining the first type of behavior and the second type of behavior in the object behavior data based on the user behavior portrait, where the first type of behavior is the behavior not in the user behavior portrait, and the second type of behavior is the behavior in the user behavior portrait; determining the first behavior score of the first type of behavior based on the behavior type of the first type of behavior, where the first behavior score is negative; determining the second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait, where the second behavior score is positive; determining the sum of the first behavior scores of each first type of behavior and the second behavior scores of each second type of behavior as the behavior score.

[0008] Optionally, determining the second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait includes: determining the proportion of the second type of behavior in the user behavior portrait; determining the second behavior score of the second type of behavior based on the proportion.

[0009] Optionally, the scoring threshold corresponding to the user behavior portrait is determined in the following manner: based on all the behavior types corresponding to the user behavior portrait, and the behavior proportion of each behavior type in the user behavior portrait; determining the Shannon diversity index of the user behavior portrait based on all the behavior types and the behavior proportion; determining the scoring threshold based on the Shannon diversity index.

[0010] Optionally, determining the scoring threshold based on the Shannon diversity index includes: determining a preset basic score and a preset adjustment coefficient; determining the ratio of the Shannon diversity index to the preset adjustment coefficient; determining the product of the ratio and the preset basic score as the scoring threshold.

[0011] Optionally, the method further includes: in the case where the behavior score is higher than the scoring threshold corresponding to the user behavior portrait, determining that the consumption behavior is not an abnormal behavior; after determining that the consumption behavior is not an abnormal behavior, determining whether the target object is the object associated with the logged-in user; in the case where the target object is determined to be the object associated with the logged-in user, updating the user behavior portrait according to the object behavior data.

[0012] According to another aspect of the embodiments of the present application, there is also provided an abnormal behavior recognition device, including: a first processing module, configured to determine object behavior data of a target object within a preset time period after detecting a consumption behavior performed by the target object on a video playback device, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; a second processing module, configured to determine a user behavior portrait corresponding to a logged-in user who logs in to the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; a third processing module, configured to determine a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; a fourth processing module, configured to determine that the consumption behavior is an abnormal behavior when the behavior score is not higher than a score threshold corresponding to the user behavior portrait.

[0013] According to another aspect of the embodiments of the present application, there is also provided a non-volatile storage medium storing a program, where, when the program runs, it controls a device where the non-volatile storage medium is located to execute an abnormal behavior recognition method.

[0014] According to another aspect of the embodiments of the present application, there is also provided an electronic device including a memory and a processor, where the processor is configured to run a program stored in the memory, and when the program runs, it executes an abnormal behavior recognition method.

[0015] According to another aspect of the embodiments of the present application, there is also provided a computer program product including a computer program, where the computer program implements an abnormal behavior recognition method when executed by a processor.

[0016] In the embodiments of the present application, after detecting a consumption behavior performed by a target object on a video playback device, object behavior data of the target object within a preset time period is determined. The object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior. A user behavior portrait corresponding to the logged-in user who logs in to the video playback device is determined. The user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device. A behavior score of the object behavior data is determined based on the user behavior portrait, and the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait. In the case where the behavior score is not higher than the score threshold corresponding to the user behavior portrait, a method for determining that the consumption behavior is an abnormal behavior is used to determine whether the consumption behavior is an abnormal behavior through the user behavior portrait, achieving the purpose of accurately identifying abnormal consumption behaviors, thereby realizing the technical effect of reducing losses for both the audience and the operator, and further solving the technical problem that the interests of both the audience and the operator are damaged due to the inability to determine whether the consumption behavior performed by the audience on the TV is an abnormal behavior in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The drawings described herein are used to provide a further understanding of the present application and form a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation to the present application. In the drawings:

[0018] Figure 1 is a schematic structural diagram of a computer terminal provided according to an embodiment of the present application;

[0019] Figure 2 is a schematic flowchart of an abnormal behavior recognition method provided according to an embodiment of the present application;

[0020] Figure 3 is a schematic flowchart of a user behavior portrait determination process provided according to an embodiment of the present application;

[0021] Figure 4 is a schematic diagram of a user behavior portrait provided according to an embodiment of the present application;

[0022] Figure 5 is a schematic flowchart of a determination process of a behavior recognition result provided according to an embodiment of the present application;

[0023] Figure 6 is a schematic flowchart of an abnormal behavior recognition process provided according to an embodiment of the present application;

[0024] Figure 7 is a schematic structural diagram of a user behavior recognition device provided according to an embodiment of the present application. Detailed Implementation Manner

[0025] To enable those skilled in the art to better understand the solution of this application, the technical solutions in the embodiments of this application will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are only a part of the embodiments of this application, rather than all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of this application.

[0026] It should be noted that the terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of this application described here can be implemented in an order other than those illustrated or described here. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or are inherent to these processes, methods, products or devices.

[0027] To better understand the embodiments of this application, the technical terms involved in the embodiments of this application are explained as follows:

[0028] IPTV (Internet Protocol Television): A technology that provides television services through Internet protocols. Different from traditional cable TV or satellite TV, IPTV transmits TV programs and video content through a broadband network. Users can watch live TV, on-demand videos, recorded programs, and use other interactive services through IPTV services. IPTV allows users to receive and watch TV programs through an Internet connection, and usually requires a dedicated set-top box or smart TV to receive the signal.

[0029] EPG (Electronic Program Guide): An interactive TV program guide that provides the schedule of TV programs and related information, such as program titles, descriptions, actor lists, etc. Users can browse different channels and programs through the EPG, select the programs they want to watch, and may set reminders or recordings. The EPG usually appears on the TV screen in the form of a graphical interface, enabling users to easily find and select the programs they are interested in. In IPTV services, the EPG is usually interactive, allowing users to directly select programs from the guide for viewing or other operations.

[0030] With the development of Internet technology, in the field of IPTV technology, the products that users can consume on the TV platform are becoming more and more abundant. Users can purchase related products according to their own needs. For example, the EPG part of the platform can provide services such as movie on demand, TV series on demand, variety show on demand, membership subscription, package subscription, etc. In addition, the TV platform also includes live shopping and video shopping. In daily life, every member of the family or visiting guest can operate the TV. It is precisely because of this feature that the operation of the TV has a natural openness. Among them, some people (such as children) may have a lack of cognition and are prone to consumption that is not in line with the actual user's wishes. For example, for children, they may consume on the TV platform at will because they have not established the correct consumption concept for the time being, which will bring unnecessary economic losses to the actual users and affect the user experience. In addition, for operators, their product sales strategies are basically formulated by humans, so there is also a risk of sales loopholes. Once the loopholes are discovered and exploited, they may cause economic losses to operators.

[0031] In order to solve the above problems, relevant solutions are provided in the embodiments of the present application, which are described in detail below.

[0032] According to an embodiment of the present application, a method embodiment of an abnormal behavior identification method is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0033] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Figure 1 FIG. 1 shows a hardware structure block diagram of a computer terminal (or mobile device) for implementing an abnormal behavior identification method. Figure 1 As shown, the computer terminal 10 (or mobile device 10) may include one or more (102a, 102b, ..., 102n are used to illustrate) processors 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA), a memory 104 for storing data, and a transmission device 106 for communication functions. In addition, it may also include: a display, an input / output interface (I / O interface), a universal serial bus (USB) port (which may be included as one of the ports of the BUS bus), a network interface, a power supply and / or a camera. It can be understood by those skilled in the art that Figure 1The structure shown is only schematic and does not limit the structure of the above-mentioned electronic device. For example, the computer terminal 10 may further include more or fewer components than those shown in Figure 1 or have a different configuration from that shown in Figure 1 .

[0034] It should be noted that one or more of the above-mentioned processors 102 and / or other data processing circuits can generally be referred to as "data processing circuits" herein. The data processing circuit can be embodied in software, hardware, firmware, or any combination thereof, in whole or in part. In addition, the data processing circuit can be a single independent processing module, or be incorporated in whole or in part into any one of the other elements in the computer terminal 10 (or mobile device). As involved in the embodiments of the present application, the data processing circuit is used for processor control (such as the selection of a variable resistance terminal path connected to an interface).

[0035] The memory 104 can be used to store software programs and modules of application software, such as the program instructions / data storage device corresponding to the abnormal behavior recognition method in the embodiments of the present application. The processor 102 executes various functional applications and data processing by running the software programs and modules stored in the memory 104, that is, implements the above-mentioned abnormal behavior recognition method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memories, or other non-volatile solid-state memories. In some instances, the memory 104 may further include a memory remotely disposed relative to the processor 102, and these remote memories can be connected to the computer terminal 10 through a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0036] The transmission device 106 is used to receive or send data via a network. Specific examples of the above-mentioned network may include a wireless network provided by the communication provider of the computer terminal 10. In one instance, the transmission device 106 includes a network adapter (Network Interface Controller, NIC), which can be connected to other network devices through a base station and thus communicate with the Internet. In one instance, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0037] The display can be, for example, a touch-screen liquid crystal display (LCD), which enables a user to interact with the user interface of the computer terminal 10 (or mobile device).

[0038] Under the above operating environment, the embodiments of the present application provide an abnormal behavior recognition method, such asFigure 2 As shown in the figure, the method includes the following steps:

[0039] Step S202: After detecting the consumption behavior performed by the target object on the video playback device, determine the object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior;

[0040] In some embodiments of the present application, various operation behavior data of the target object within the first time period will be stored in the first storage space. When it is detected that the target object has performed a consumption behavior, the implementation time point of the consumption behavior can be used as the end point of the preset time period, and the object behavior data of the preset duration can be retrieved and obtained forward, so as to obtain the behavior data within the preset time period. Various types of behavior data include operation behavior data, viewing behavior data, consumption behavior data, etc.

[0041] As an alternative implementation, when the user consumes products on the TV or continuously consumes products, regardless of the amount of consumption, the monitoring mechanism will be triggered to obtain the current user behavior data before and after the trigger time. And use the clustering clusters used during the construction of the portrait to classify the operation behavior, viewing behavior, and consumption behavior in the current user behavior data, so as to obtain the types of operation behavior, viewing behavior, and consumption behavior of user A at this time.

[0042] Step S204: Determine the user behavior portrait corresponding to the logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device;

[0043] In the technical solution provided in step S204, the user behavior portrait includes an operation behavior portrait, a viewing behavior portrait, and a consumption behavior portrait; the user behavior portrait can be generated in the following manner: clustering the behavior data of multiple users to obtain the types of operation behavior, viewing behavior, and consumption behavior, where the behavior data includes operation behavior data, viewing behavior data, and consumption behavior data; obtaining the historical user data of the logged-in user, and determining the user behavior portrait based on the historical user data and the types of operation behavior, viewing behavior, and consumption behavior, where the historical user data includes historical operation behavior data, historical viewing behavior data, and historical consumption behavior data.

[0044] In some embodiments of the present application, the process of constructing the user behavior portrait is as Figure 3 shown. From Figure 3It can be seen that the behavior data of different TV users can be obtained through the probes set in the set-top box, and the collected behavior data will also be transmitted to the cloud platform. Then, clustering and classification operations are performed on the user behavior on the cloud platform, and a portrait of the user is constructed according to the classification result of the user behavior. The constructed portrait of the user is used to judge the abnormal behavior of the user.

[0045] Optionally, the behavior portrait can include three parts: operation behavior portrait, viewing behavior portrait, and consumption behavior portrait. It is mainly constructed from the user data obtained by collecting the historical behaviors of users through the probes on the TV. For example, the user data can include: the operation data of the user, and the operation data of the user is the jump path of the TV interface when controlling the set-top box and the operation duration when the interface jump is completed. The program viewing records of the user, such as the type of the viewed program, the viewing duration of the program, the time period of viewing the program, etc. To ensure the validity of the key program records, after the data is collected, the data needs to be screened, and the screening is mainly to remove the program viewing records with too short viewing duration. The consumption records of the user, such as the amount of consumption, the content of consumption, and the consumption frequency, etc. After the user data is collected, the user data is then cleaned.

[0046] The operation behavior portrait corresponds to the path of the interface jump on the TV when controlling the set-top box, the viewing behavior portrait corresponds to the user's program viewing records, and the consumption behavior portrait corresponds to the user's consumption records. When constructing the user behavior portrait, a large amount of user data from different users can be used for clustering first to obtain the types of operation behavior, viewing behavior, and consumption behavior respectively. Taking the operation behavior as an example, the K-Means algorithm is used to classify the operation data in the user data (the operation data can be first converted into sample points in a multi-dimensional space and then clustered, and different features in the operation data correspond to different dimensions in the multi-dimensional space), forming different clusters, and one cluster is one kind of operation behavior. Similarly, the types of viewing behavior and consumption behavior are determined through clustering respectively.

[0047] Then, for user A, in order to obtain the portrait of user A, the behaviors in the historical behavior data of user A can be classified according to the clustering result, so as to obtain the operation behavior portrait, viewing behavior portrait, and consumption behavior portrait of user A. As Figure 4 shown, the operation behavior portrait includes the distribution of the types of operation behaviors of user A, the viewing behavior portrait includes the distribution of the types of viewing behaviors of user A, and the consumption behavior portrait includes the distribution of the types of consumption behaviors.

[0048] As an optional implementation manner, as Figure 5 shown, the following process can also be adopted to identify the types of various behaviors included in the behavior data of the target object:

[0049] For any clustering group in the operation behavior part, the cluster center of the clustering group can be determined based on each behavior sample point in the clustering group (i.e., the result after data transformation into a multi-dimensional space). The cluster center can be the central position of the clustering group or the central sample closest to the central position. Among them, the cluster center can be obtained by taking the average of each behavior sample point in the clustering group.

[0050] After determining the cluster center of each clustering group, then calculate the distance between the point after converting the operation behavior part in the current user behavior data into the corresponding multi-dimensional space and each cluster center, and take the clustering group where the closest cluster center is located as the attribution cluster of the above current user behavior data, so as to obtain the type of operation behavior of the target object.

[0051] Through the above method, the types of operation behavior, viewing behavior, and consumption behavior of the target object can be obtained.

[0052] Step S206: Determine the behavior score of the object behavior data according to the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait;

[0053] After obtaining the recognition result of the behavior type of the target object, the recognition result can be further compared with the user behavior portrait of user A (i.e., the currently logged-in user), so as to determine the score of the object behavior data, and further determine whether the consumption behavior is an abnormal consumption behavior.

[0054] In the technical solution provided in step S206, the steps of determining the behavior score of the object behavior data according to the user behavior portrait include: determining the first type of behavior and the second type of behavior in the object behavior data according to the user behavior portrait, where the first type of behavior is the behavior not in the user behavior portrait, and the second type of behavior is the behavior in the user behavior portrait; determining the first behavior score of the first type of behavior according to the behavior type of the first type of behavior, where the first behavior score is negative; determining the second behavior score of the second type of behavior according to the distribution of the second type of behavior in the user behavior portrait, where the second behavior score is positive; determining the sum of the first behavior scores of each first type of behavior and the second behavior scores of each second type of behavior as the behavior score.

[0055] In some embodiments of the present application, the steps of determining the second behavior score of the second type of behavior according to the distribution of the second type of behavior in the user behavior portrait include: determining the proportion of the second type of behavior in the user behavior portrait; determining the second behavior score of the second type of behavior according to the proportion.

[0056] Optionally, when scoring, the score of a certain behavior can be obtained according to the distribution of various behavior types in the user behavior portrait, and then the behavior score of user A can be obtained by synthesizing the scores of each behavior. Compare the behavior score of the user with the score threshold corresponding to the abnormal behavior to determine whether it is an abnormal behavior.

[0057] The scoring method for each user behavior is as follows:

[0058] If it is recognized that the target object has user behavior x and there is no user behavior x in the user behavior portrait, a negative score is directly assigned to user behavior x. (The negative scores assigned to different user behaviors x can be different. For example, for more extreme user behaviors x, when it does not exist in the user portrait, the assigned negative score is smaller);

[0059] If it is recognized that the target object has user behavior x and there is user behavior x in the user behavior portrait, the score of user behavior x is calculated according to the proportion of user behavior x in the user behavior portrait, and the score of user behavior x is directly proportional to the proportion it occupies. For example, multiplying the proportion it occupies by a coefficient can obtain the score.

[0060] The score calculation formula for a single user behavior is as follows:

[0061]

[0062] In the formula, Ai is the assignment result of behavior i, and its value is less than or equal to 0; Pj is the proportion of behavior j in the user portrait; α is the coefficient of the behavior type and is a positive number.

[0063] The comprehensive score calculation formula for a certain user is as follows:

[0064]

[0065] In the formula, M s is the comprehensive score of a certain user, is the cumulative score of the behaviors that do not belong to the user portrait, is the cumulative score of the behaviors that belong to the user portrait.

[0066] Step S208, when the behavior score is not higher than the score threshold corresponding to the user behavior portrait, determine that the consumption behavior is an abnormal behavior.

[0067] In the technical solution provided in step S208, the score threshold corresponding to the user behavior portrait is determined in the following manner: based on all the behavior types corresponding to the user behavior portrait and the behavior proportions of each behavior type in the user behavior portrait; determine the Shannon diversity index of the user behavior portrait based on all the behavior types and behavior proportions; determine the score threshold based on the Shannon diversity index.

[0068] As an alternative implementation, the steps of determining the scoring threshold based on the Shannon diversity index include: determining a preset base score and a preset adjustment coefficient; determining the ratio of the Shannon diversity index to the preset adjustment coefficient; and determining the product of the ratio and the preset base score as the scoring threshold.

[0069] Optionally, for the score threshold corresponding to abnormal behavior, different score thresholds can be formed for different users. For example, for users with variable behaviors, the corresponding score threshold will be higher, while for users with stable behaviors, the corresponding score threshold will be lower. In this way, since the score threshold for users with variable behaviors is higher, when determining whether there is abnormal consumption behavior in the user behavior portrait applicable to such users, the sensitivity of the judgment process can be improved, and it is easier to determine whether there is abnormal consumption behavior. Specifically as follows:

[0070] For user A, after generating the behavior portrait of user A, determine the score threshold of user A through the distribution of each behavior in the behavior portrait. For example, the Shannon diversity index can be used to characterize the number of behavior types and the concentration degree of behaviors in the portrait, and the score threshold can be calculated in combination with the Shannon diversity index.

[0071] The Shannon diversity index is:

[0072]

[0073] In the formula, H is the Shannon diversity index, S is the total number of behavior types in the portrait, and pi is the proportion of the number corresponding to behavior i in the portrait.

[0074] The Shannon diversity index H is proportional to the number of behavior types and also proportional to the uniformity of the distribution of individuals among the types. Therefore, a portrait with a large number of behavior types and a uniform behavior distribution will have a higher Shannon diversity index, and such a portrait can correspond to the above-mentioned users with variable behaviors.

[0075] The calculation formula for the score threshold is as follows:

[0076]

[0077] In the formula, M y is the score threshold, M l is the base score (which can be set manually, such as 5 points), H is the Shannon diversity index, and γ is the adjustment coefficient (which can be set manually, such as 0.8).

[0078] Finally, compare M s with M y in terms of size. If M s is less than or equal to M y, it can be considered that the behavior of the current user is abnormal, and the consumption behavior of the user can be interfered, such as prohibiting the current consumption behavior or sending a prompt confirmation message.

[0079] Suppose there are two sets of data samples as follows:

[0080]

[0081] As shown in the above table, Group 1 has more types and a more dispersed behavior distribution, while Group 2 has fewer types and a more concentrated behavior distribution. Correspondingly, Group 1 has a higher score threshold of 8.68.

[0082] In some embodiments of the present application, the user behavior portrait can also be updated in the following manner: when the behavior score is higher than the score threshold corresponding to the user behavior portrait, it is determined that the consumption behavior is not an abnormal behavior; after determining that the consumption behavior is not an abnormal behavior, it is determined whether the target object is an object associated with the logged-in user; when it is determined that the target object is an object associated with the logged-in user, the user behavior portrait is updated according to the object behavior data.

[0083] In some embodiments of the present application, the process of identifying abnormal behavior is as Figure 6 shown, including steps such as establishing a user behavior portrait, detecting the real-time behavior of the user, identifying and determining whether it is an abnormal behavior, and determining whether to intervene in the user behavior according to the identification result. The real-time behavior of the user detected is not necessarily the behavior of the user corresponding to the user behavior portrait. For example, if the logged-in user of the current TV is A, and the user watching the TV at this time is B, then the user portrait used in the present application is the user behavior portrait of user A, and the user behavior obtained and identified is the behavior of user B.

[0084] For example, when user B brings a child to visit user A's home and allows the child to watch TV on their own. Due to the child's incomplete cognition or other reasons, the child consumes the products on TV without supervision, triggering the monitoring mechanism in the present invention. The system automatically obtains the current behavior data before and after the child's actions and classifies them to obtain the types of the child's operation behaviors, viewing behaviors, and consumption behaviors. Then, it compares with user A's portrait. Suppose the classified child behaviors are operation behavior a1, viewing behavior b2, and consumption behavior a1. Among them, viewing behavior b2 and consumption behavior a1 do not belong to user A's portrait. When calculating the score of a single user behavior, viewing behavior b2 and consumption behavior a1 are respectively assigned -2 and -5. If the proportion of operation behavior a1 is 20%, the score of operation behavior a1 is calculated as 20·α(0.15) to get 3. The cumulative comprehensive score M_s of the child's behaviors = 3 - 2 - 5 = -4. And the score threshold calculated through user A's portrait is 7 (the score threshold can be updated synchronously after the user portrait is updated and does not need to be calculated every time a behavior is recognized). After that, since the comprehensive score M_s of the child's behaviors is less than 7, it can be determined that the behavior is abnormal at this time. And timely intervention actions are taken to restrict the child's consumption to avoid unnecessary losses to user A and affect user A's usage experience. In addition, if user A himself / herself has children and user A often consumes products (such as paid animations) for his / her own children on TV, then viewing behavior b2 and consumption behavior a1 in the above child's behaviors may both exist in user A's portrait. Therefore, even if consumption occurs in this case, it will not finally be determined as an abnormal behavior. Because the above child's behavior conforms to user A's tendency and will not affect user A's usage experience, it can be not determined as an abnormal behavior.

[0085] As an optional implementation manner, in order to distinguish whether the consumption behavior is implemented by user A or the guest when user A himself / herself has children, timestamp information can be added to the consumption behavior. And when a consumption behavior is recognized, determine whether the occurrence time point of the consumption behavior is within the time period of user A's common consumption behaviors. If not, issue a prompt or interfere with the consumption behavior to avoid losses to user A.

[0086] In some embodiments of the present application, assuming that there is a problem (or a bug) in the operator's promotion mechanism, resulting in the actual amount that the user needs to pay being much less than the operator's promotion price, it may trigger frequent consumption behaviors of the user. At this time, the user's behavior data and consumption behaviors may deviate significantly from the user's original portrait. For example, there may be repeated and frequent operations, high-frequency consumption, and new consumption content. At this time, the types of user behavior data and consumption behaviors identified basically do not exist in the user's portrait. Correspondingly, the comprehensive score M_s of the user calculated will be less than M_y, and the user's behavior is identified as abnormal, and consumption is restricted in a timely manner to avoid unnecessary losses to the TV product operator.

[0087] By determining the object behavior data of the target object within a preset time period after detecting the consumption behavior performed by the target object on the video playback device, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; determining the user behavior portrait corresponding to the logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; determining the behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and in the case where the behavior score is not higher than the score threshold corresponding to the user behavior portrait, determining that the consumption behavior is an abnormal behavior. By using the user behavior portrait to determine whether the consumption behavior is an abnormal behavior, the purpose of accurately identifying abnormal consumption behaviors is achieved, thereby realizing the technical effect of reducing losses for both the audience and the operator, and further solving the technical problem that the interests of both the audience and the operator are damaged due to the inability to determine whether the consumption behavior performed by the audience on the TV is an abnormal consumption behavior in the related art.

[0088] The embodiments of the present application provide an abnormal behavior recognition device. Figure 7 is a schematic structural diagram of the device. From Figure 7As can be seen, the device includes: a first processing module 70, configured to determine object behavior data of a target object within a preset time period after detecting a consumption behavior performed by the target object on a video playback device, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; a second processing module 72, configured to determine a user behavior portrait corresponding to a logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; a third processing module 74, configured to determine a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and a fourth processing module 76, configured to determine that the consumption behavior is an abnormal behavior when the behavior score is not higher than a score threshold corresponding to the user behavior portrait.

[0089] In some embodiments of the present application, the user behavior portrait includes an operation behavior portrait, a viewing behavior portrait, and a consumption behavior portrait; the second processing module 72 is further configured to: cluster the behavior data of multiple users to obtain operation behavior types, viewing behavior types, and consumption behavior types, where the behavior data includes operation behavior data, viewing behavior data, and consumption behavior data; obtain the historical user data of the logged-in user, and determine the user behavior portrait based on the historical user data and the operation behavior types, viewing behavior types, and consumption behavior types, where the historical user data includes historical operation behavior data, historical viewing behavior data, and historical consumption behavior data.

[0090] In some embodiments of the present application, the steps for the third processing module 74 to determine the behavior score of the object behavior data based on the user behavior portrait include: determining a first type of behavior and a second type of behavior in the object behavior data based on the user behavior portrait, where the first type of behavior is a behavior not in the user behavior portrait, and the second type of behavior is a behavior in the user behavior portrait; determining a first behavior score of the first type of behavior based on the behavior type of the first type of behavior, where the first behavior score is negative; determining a second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait, where the second behavior score is positive; and determining the sum of the first behavior scores of each first type of behavior and the second behavior scores of each second type of behavior as the behavior score.

[0091] In some embodiments of the present application, the steps for the third processing module 74 to determine the second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait include: determining the proportion of the second type of behavior in the user behavior portrait; and determining the second behavior score of the second type of behavior based on the proportion.

[0092] In some embodiments of the present application, the scoring threshold corresponding to the user behavior portrait is determined in the following manner: based on all the behavior types corresponding to the user behavior portrait and the behavior proportion of each behavior type in the user behavior portrait; determining the Shannon diversity index of the user behavior portrait based on all the behavior types and the behavior proportion; and determining the scoring threshold based on the Shannon diversity index.

[0093] In some embodiments of the present application, the steps for the fourth processing module 76 to determine the scoring threshold based on the Shannon diversity index include: determining a preset basic score and a preset adjustment coefficient; determining the ratio of the Shannon diversity index to the preset adjustment coefficient; and determining the product of the ratio and the preset basic score as the scoring threshold.

[0094] In some embodiments of the present application, the fourth processing module 76 is further configured to: when the behavior score is higher than the scoring threshold corresponding to the user behavior portrait, determine that the consumption behavior is not an abnormal behavior; after determining that the consumption behavior is not an abnormal behavior, determine whether the target object is an object associated with the logged-in user; and when determining that the target object is an object associated with the logged-in user, update the user behavior portrait according to the object behavior data.

[0095] It should be noted that each module in the above abnormal behavior recognition device may be a program module (for example, a set of program instructions for implementing a specific function), or a hardware module. For the latter, it may be presented in the following forms, but not limited to: the manifestation form of each of the above modules is a processor, or the functions of each of the above modules are implemented by a processor.

[0096] According to an embodiment of the present application, there is also provided a non-volatile storage medium. A program is stored in the non-volatile storage medium. When the program runs, it controls the device where the non-volatile storage medium is located to execute the following abnormal behavior recognition method: after detecting the consumption behavior performed by the target object on the video playback device, determine the object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; determine the user behavior portrait corresponding to the logged-in user who logs in to the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; determine the behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and when the behavior score is not higher than the scoring threshold corresponding to the user behavior portrait, determine that the consumption behavior is an abnormal behavior.

[0097] According to an embodiment of the present application, an electronic device is further provided, including a memory and a processor. The processor is used to run a program stored in the memory. When the program runs, the following abnormal behavior recognition method is executed: after detecting a consumption behavior performed by a target object on a video playback device, determining object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; determining a user behavior portrait corresponding to the logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; determining a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and when the behavior score is not higher than the score threshold corresponding to the user behavior portrait, determining that the consumption behavior is an abnormal behavior.

[0098] According to an embodiment of the present application, a computer program product is further provided, including a computer program. When the computer program is executed by a processor, the following abnormal behavior recognition method is implemented: after detecting a consumption behavior performed by a target object on a video playback device, determining object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; determining a user behavior portrait corresponding to the logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; determining a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; and when the behavior score is not higher than the score threshold corresponding to the user behavior portrait, determining that the consumption behavior is an abnormal behavior.

[0099] In the above embodiments of the present application, the descriptions of the various embodiments have their own focuses. For parts not detailed in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0100] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are merely illustrative. For example, the division of the units can be a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of units or modules can be in an electrical or other form.

[0101] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed over multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0102] In addition, each functional unit in various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0103] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the related technology, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), mobile hard disks, magnetic disks or optical discs that can store program codes.

[0104] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

Claims

1. An abnormal behavior recognition method, characterized in that, Including: After detecting a consumption behavior performed by a target object on a video playback device, determining object behavior data of the target object within a preset time period, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined based on the implementation time point of the consumption behavior; Determining a user behavior portrait corresponding to a logged-in user of the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; Determining a behavior score of the object behavior data based on the user behavior portrait, where the behavior score is used to reflect the similarity degree between the behavior data and the user behavior portrait; When the behavior score is not higher than a score threshold corresponding to the user behavior portrait, determining that the consumption behavior is an abnormal behavior.

2. The abnormal behavior recognition method according to claim 1, characterized in that The user behavior portrait includes an operation behavior portrait, a viewing behavior portrait, and a consumption behavior portrait; the method further includes: Clustering the behavior data of multiple users to obtain operation behavior types, viewing behavior types, and consumption behavior types, where the behavior data includes operation behavior data, viewing behavior data, and consumption behavior data; Obtaining historical user data of the logged-in user, and determining the user behavior portrait based on the historical user data, the operation behavior types, the viewing behavior types, and the consumption behavior types, where the historical user data includes historical operation behavior data, historical viewing behavior data, and historical consumption behavior data.

3. The abnormal behavior recognition method according to claim 1, characterized in that, Determining the behavior score of the object behavior data based on the user behavior portrait includes: Determining a first type of behavior and a second type of behavior in the object behavior data based on the user behavior portrait, where the first type of behavior is a behavior not in the user behavior portrait, and the second type of behavior is a behavior in the user behavior portrait; Determining a first behavior score of the first type of behavior based on the behavior type of the first type of behavior, where the first behavior score is negative; Determining a second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait, where the second behavior score is positive; Determining the sum of the first behavior scores of each of the first type of behaviors and the second behavior scores of each of the second type of behaviors as the behavior score.

4. The abnormal behavior recognition method according to claim 3, wherein Determining the second behavior score of the second type of behavior based on the distribution of the second type of behavior in the user behavior portrait includes: Determining the proportion of the second type of behavior in the user behavior portrait; Determining the second behavior score of the second type of behavior based on the proportion.

5. The abnormal behavior recognition method according to claim 1, characterized in that, The score threshold corresponding to the user behavior portrait is determined by the following method: Based on all behavior types corresponding to the user behavior portrait, and the behavior proportion of each behavior type in the user behavior portrait; Determining the Shannon diversity index of the user behavior portrait based on all the behavior types and the behavior proportion; Determining the score threshold based on the Shannon diversity index.

6. The abnormal behavior recognition method according to claim 5, wherein Determining the scoring threshold according to the Shannon diversity index includes: Determining a preset basic score and a preset adjustment coefficient; Determining the ratio of the Shannon diversity index to the preset adjustment coefficient; Determining the product of the ratio and the preset basic score as the scoring threshold.

7. The abnormal behavior recognition method according to claim 1, wherein The method further includes: When the behavior score is higher than the scoring threshold corresponding to the user behavior portrait, determining that the consumption behavior is not an abnormal behavior; After determining that the consumption behavior is not an abnormal behavior, determining whether the target object is an object associated with the logged-in user; When it is determined that the target object is an object associated with the logged-in user, updating the user behavior portrait according to the object behavior data.

8. An abnormal behavior recognition device, characterized in that, Including: A first processing module, configured to determine object behavior data of the target object within a preset time period after detecting a consumption behavior performed by the target object on a video playback device, where the object behavior data includes various behaviors implemented by the target object through the video playback device within the preset time period, and the preset time period is a time period determined according to the implementation time point of the consumption behavior; A second processing module, configured to determine a user behavior portrait corresponding to a logged-in user who logs in to the video playback device, where the user behavior portrait includes the distribution of various behaviors implemented by the logged-in user through the video playback device; A third processing module, configured to determine a behavior score of the object behavior data according to the user behavior portrait, where the behavior score is used to reflect the similarity between the behavior data and the user behavior portrait; A fourth processing module, configured to determine that the consumption behavior is an abnormal behavior when the behavior score is not higher than the scoring threshold corresponding to the user behavior portrait.

9. A non-volatile storage medium, characterized in that, The non-volatile storage medium stores a program, where, when the program runs, it controls the device where the non-volatile storage medium is located to execute the abnormal behavior recognition method according to any one of claims 1 to 7.

10. An electronic device, characterized in that, Including: A memory and a processor, where the processor is configured to run a program stored in the memory, and when the program runs, it executes the abnormal behavior recognition method according to any one of claims 1 to 7.

11. A computer program product, characterized in that, Including a computer program, where the computer program, when executed by a processor, implements the abnormal behavior recognition method according to any one of claims 1 to 7.

Images