Identification method, device and equipment, computer readable storage medium and program product

Abstract

The embodiment of the invention provides a recognition method, device and equipment, a computer readable storage medium and a program product, and relates to the fields of artificial intelligence, maps and the like, and application scenes include but are not limited to quantity refreshing behavior scenes. The method comprises the following steps: based on operation data of a plurality of to-be-identified users, determining a user level of each to-be-identified user in the plurality of to-be-identified users and a user group to which the to-be-identified user belongs; determining a plurality of operation behavior segments based on the operation data of the plurality of to-be-identified users; based on each user group and each operation behavior segment, determining a plurality of initial target operation behavior segments and a to-be-identified user set corresponding to the plurality of initial target operation behavior segments; based on the multiple initial target operation behavior segments and the to-be-recognized user set, determining a relation graph; and based on the relation graph, determining a target operation behavior segment from the multiple initial target operation behavior segments, and based on the to-be-identified user corresponding to the target operation behavior segment, determining the user with the amount refreshing behavior and the group with the amount refreshing behavior.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and more specifically, to an identification method, apparatus, device, computer-readable storage medium, and program product. Background Art

[0002] In a video APP (Application), some users will create and publish some videos, and other users will pay attention to these videos and browse, like, comment, forward, collect, etc. However, in order to quickly attract attention to the video work, a video author will purchase a volume-boosting service. After purchasing the volume-boosting service, a certain volume-boosting group will browse, like, comment, etc. on the video work of the video author in batches, so that the video work will generate a lot of popularity in a short period of time. Among them, the volume-boosting group is the group that has volume-boosting behavior; in this way, the traffic platform may determine that the video work is a high-quality video and recommend the video work to a large number of users, so that the video work will receive a lot of exposure and attention, so that the video author can benefit from it.

[0003] In the prior art, if a user performs a large number of operations, the user is determined to be a brushing user, and a brushing user is a user who engages in brushing behavior; however, normal users can also perform a large number of operations. Therefore, determining that a user is a brushing user based solely on the fact that the user performs a large number of operations often leads to misjudgment, thereby reducing the accuracy of identifying users who engage in brushing behavior, that is, reducing the accuracy of identifying groups that engage in brushing behavior. Summary of the invention

[0004] In view of the shortcomings of existing methods, the present application proposes an identification method, device, equipment, computer-readable storage medium and computer program product to solve the problem of how to improve the accuracy of identifying groups that engage in brushing behavior.

[0005] In a first aspect, the present application provides an identification method, comprising:

[0006] Obtaining operation data of multiple users to be identified;

[0007] Based on the operation data of the multiple users to be identified, determine the user level of each user to be identified and the user group to which the user belongs through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group;

[0008] Based on the operation data of multiple users to be identified, multiple operation behavior segments are determined through segmentation and classification processing, each of the multiple operation behavior segments includes multiple operation behaviors arranged in chronological order;

[0009] Based on each user group and each operation behavior fragment, multiple initial target operation behavior fragments and sets of users to be identified corresponding to the multiple initial target operation behavior fragments are determined through evaluation and screening processing;

[0010] Based on the multiple initial target operation behavior fragments and the set of users to be identified, a relationship graph is determined, wherein the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or any one of the users to be identified in the set of users to be identified;

[0011] Based on the relationship graph, through node importance evaluation processing, the target operation behavior fragment is determined from multiple initial target operation behavior fragments, and based on the to-be-identified users corresponding to the target operation behavior fragment, the users and groups that have engaged in brushing behavior are determined.

[0012] In one embodiment, based on the operation data of the plurality of users to be identified, determining the user level of each of the plurality of users to be identified through weighted processing includes:

[0013] For each of the multiple users to be identified, based on the operation data of each user to be identified within a preset time period, determining multiple operation behaviors of each user to be identified and the number of operations corresponding to each of the multiple operation behaviors through classification processing;

[0014] Based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors, an operation behavior index corresponding to each to-be-identified user is determined through weighted summation processing;

[0015] Based on the operation behavior index corresponding to each user to be identified and the preset classification threshold, classification processing is performed to determine the user level of each user to be identified.

[0016] In one embodiment, based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors, the operation behavior index corresponding to each to-be-identified user is determined through weighted summation processing, including:

[0017] The number of operations corresponding to each operation behavior in various operation behaviors and the preset weight corresponding to each operation behavior are multiplied to obtain a weighted value corresponding to each operation behavior;

[0018] The weighted values ​​corresponding to various operation behaviors are summed up to determine the operation behavior index corresponding to each user to be identified.

[0019] In one embodiment, based on the operation data of multiple users to be identified, multiple operation behavior segments are determined through segmentation and classification, including:

[0020] For each of the multiple users to be identified, based on the operation data of each of the users to be identified, determining an operation behavior sequence of each of the users to be identified;

[0021] Based on the operation behavior sequence of each to-be-identified user, the preset operation behavior segment length, and the preset sliding window length, segmentation processing is performed to determine multiple operation behavior segments of each to-be-identified user, each of the multiple operation behavior segments including multiple operation behaviors performed by each to-be-identified user in chronological order;

[0022] Based on the multiple operation behavior fragments of each to-be-identified user, classification processing is performed to determine the multiple operation behavior fragments corresponding to each to-be-identified user.

[0023] In one embodiment, based on each user group and each operation behavior segment, multiple initial target operation behavior segments and user sets to be identified corresponding to the multiple initial target operation behavior segments are determined through evaluation and screening, including:

[0024] Based on each user group and each operation behavior segment, determining the growth trend value of each operation behavior segment through growth trend evaluation processing;

[0025] Based on the growth trend values ​​of various operation behavior segments, a plurality of initial target operation behavior segments and sets of users to be identified corresponding to the plurality of initial target operation behavior segments are determined through screening processing.

[0026] In one embodiment, based on each user group and each operation behavior segment, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing, including:

[0027] Based on each user group and each operation behavior segment, determine the proportion of each operation behavior segment in each user group through ratio calculation processing;

[0028] Based on the proportion of each operation behavior segment in each user group, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing.

[0029] In one embodiment, based on the proportion of each operation behavior segment in each user group, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing, including:

[0030] Based on the proportion of each operation behavior segment in each user group, through proportion calculation processing, determine the proportion continuous growth judgment value corresponding to each operation behavior segment, the proportion continuous growth amplitude corresponding to each operation behavior segment and the proportion continuous growth multiple corresponding to each operation behavior segment. The proportion continuous growth judgment value is used to indicate whether the proportion of each operation behavior segment in each user group is proportional to the user level corresponding to each user group.

[0031] Based on the judgment value of continuous growth in proportion, the amplitude of continuous growth in proportion and the multiple of continuous growth in proportion, the growth trend value of each operation behavior segment is determined through multiplication processing.

[0032] In one embodiment, based on the growth trend values ​​of various operation behavior segments, multiple initial target operation behavior segments and user sets to be identified corresponding to the multiple initial target operation behavior segments are determined through screening processing, including:

[0033] If the growth trend value of any one of the various operation behavior segments is greater than a preset growth trend threshold, then determining any one of the operation behavior segments as an initial target operation behavior segment;

[0034] The to-be-identified users corresponding to various initial target operation behavior fragments are constructed into a to-be-identified user set corresponding to various initial target operation behavior fragments.

[0035] In one embodiment, based on the relationship graph, a target operation behavior segment is determined from a plurality of initial target operation behavior segments through node importance evaluation processing, including:

[0036] Based on the relationship graph, the node importance corresponding to any one of the multiple initial target operation behavior segments is determined through node importance evaluation processing, and the node importance corresponding to any one of the initial target operation behavior segments is the stable probability value of the node in the relationship graph corresponding to any one of the initial target operation behavior segments;

[0037] If the node importance corresponding to any initial target operation behavior segment is greater than a preset importance threshold, then any initial target operation behavior segment is determined as a target operation behavior segment.

[0038] In one embodiment, based on the to-be-identified user corresponding to the target operation behavior segment, determining the user who has engaged in the brushing behavior and the group who has engaged in the brushing behavior includes:

[0039] For any user to be identified corresponding to the target operation behavior segment, if the number of times any user to be identified performs the target operation behavior segment is greater than a preset number threshold, then any user to be identified is determined as a user who has engaged in brushing behavior;

[0040] Based on the users who have engaged in inflating traffic, the group that has engaged in inflating traffic is determined through statistical estimation.

[0041] In one embodiment, based on each user who has engaged in the inflating behavior, a group that has engaged in the inflating behavior is determined through statistical estimation processing, including:

[0042] If it is determined that multiple users who have engaged in brushing behavior have engaged in brushing behavior on the same work, the same work will be determined as a brushed work, and the multiple users who have engaged in brushing behavior will be determined as a group that has engaged in brushing behavior.

[0043] In a second aspect, the present application provides an identification device, comprising:

[0044] A first processing module is used to obtain operation data of multiple users to be identified;

[0045] The second processing module is used to determine the user level of each of the multiple users to be identified and the user group to which the user belongs through weighted processing based on the operation data of the multiple users to be identified, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group;

[0046] A third processing module is used to determine multiple operation behavior segments based on the operation data of multiple users to be identified through segmentation and classification processing, wherein each of the multiple operation behavior segments includes multiple operation behaviors arranged in chronological order;

[0047] A fourth processing module is used to determine, based on each user group and each operation behavior segment, a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through evaluation and screening processing;

[0048] a fifth processing module, configured to determine a relationship graph based on the multiple initial target operation behavior segments and the set of users to be identified, wherein the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior segments or any one of the users to be identified in the set of users to be identified;

[0049] The sixth processing module is used to determine the target operation behavior fragment from multiple initial target operation behavior fragments through node importance evaluation processing based on the relationship graph, and determine the users and groups that have engaged in brushing behavior based on the users to be identified corresponding to the target operation behavior fragment.

[0050] In a third aspect, the present application provides an electronic device, including: a processor, a memory, and a bus;

[0051] A bus, used to connect the processor and memory;

[0052] A memory, used for storing operation instructions;

[0053] The processor is used to execute the identification method of the first aspect of the present application by calling an operation instruction.

[0054] In a fourth aspect, the present application provides a computer-readable storage medium storing a computer program, and the computer program is used to execute the identification method of the first aspect of the present application.

[0055] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the steps of the identification method in the first aspect of the present application when the computer program is executed by a processor.

[0056] The technical solution provided in the embodiments of the present application has at least the following beneficial effects:

[0057] Acquire operation data of multiple users to be identified; determine the user level of each user to be identified and the user group to which the user belongs, based on the operation data of the multiple users to be identified, through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group; determine multiple operation behavior fragments based on the operation data of the multiple users to be identified, through segmentation and classification processing, each of the multiple operation behavior fragments includes multiple operation behaviors arranged in chronological order; determine multiple initial target operation behavior fragments and the user sets to be identified corresponding to the multiple initial target operation behavior fragments through evaluation and screening processing based on each user group and each operation behavior fragment; determine a relationship graph based on the multiple initial target operation behavior fragments and the user set to be identified, the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or any one of the user sets to be identified. Identify users; based on the relationship graph, determine the target operation behavior fragment from multiple initial target operation behavior fragments through node importance evaluation processing, and based on the users to be identified corresponding to the target operation behavior fragment, determine the users who have brushed the volume and the groups who have brushed the volume; in this way, determine the user level of each user to be identified, the higher the user level, the more times the user to be identified has performed brushing behaviors, and the more times the brushing behaviors are performed, the greater the probability that the user to be identified is a brushing user; based on user groups with different user levels and each operation behavior fragment, identify multiple initial target operation behavior fragments (suspected brushing operation behavior fragments) and sets of users to be identified (suspected brushing user sets) corresponding to the multiple initial target operation behavior fragments; based on the multiple initial target operation behavior fragments and the set of users to be identified, construct a relationship graph, and based on the relationship graph, determine the users who have brushed the volume and the groups who have brushed the volume, thereby improving the accuracy of identifying the groups who have brushed the volume. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in describing the embodiments of the present application are briefly introduced below.

[0059] Figure 1 A schematic diagram of the architecture of the identification system provided in the embodiment of the present application;

[0060] Figure 2 A flowchart of an identification method provided in an embodiment of the present application;

[0061] Figure 3 A schematic diagram of identification provided for an embodiment of the present application;

[0062] Figure 4 A schematic diagram of identification provided for an embodiment of the present application;

[0063] Figure 5 A schematic diagram of identification provided for an embodiment of the present application;

[0064] Figure 6 A flowchart of an identification method provided in an embodiment of the present application;

[0065] Figure 7 A schematic diagram of the brushing process provided in an embodiment of the present application;

[0066] Figure 8 A schematic diagram of the structure of an identification device provided in an embodiment of the present application;

[0067] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0068] The embodiments of the present application are described below in conjunction with the drawings in the present application. It should be understood that the implementation methods described below in conjunction with the drawings are exemplary descriptions for explaining the technical solutions of the embodiments of the present application and do not constitute a limitation on the technical solutions of the embodiments of the present application.

[0069] It will be understood by those skilled in the art that, unless specifically stated, the singular forms "one", "said", and "the" used herein may also include plural forms. It should be further understood that the terms "including" and "comprising" used in the embodiments of the present application refer to that the corresponding features can be implemented as the presented features, information, data, steps, operations, elements and / or components, but do not exclude the implementation as other features, information, data, steps, operations, elements, components and / or combinations thereof supported by the technical field. It should be understood that when we say that an element is "connected" or "coupled" to another element, the one element may be directly connected or coupled to the other element, or it may refer to that the one element and the other element establish a connection relationship through an intermediate element. In addition, the "connection" or "coupling" used herein may include wireless connection or wireless coupling. The term "and / or" used herein indicates at least one of the items defined by the term, for example, "A and / or B" indicates that it is implemented as "A", or implemented as "B", or implemented as "A and B".

[0070] It is understandable that in the specific implementation of this application, it involves identifying relevant data. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0071] In order to make the objectives, technical solutions and advantages of the present application clearer, the implementation methods of the present application will be further described in detail below with reference to the accompanying drawings.

[0072] The embodiment of the present application is a recognition method provided by a recognition system, which relates to fields such as artificial intelligence and maps.

[0073] Artificial Intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results. In other words, artificial intelligence is a comprehensive technology in computer science that attempts to understand the essence of intelligence and produce a new intelligent machine that can respond in a similar way to human intelligence. Artificial intelligence is to study the design principles and implementation methods of various intelligent machines so that machines have the functions of perception, reasoning and decision-making.

[0074] Artificial intelligence technology is a comprehensive discipline that covers a wide range of fields, including both hardware-level and software-level technologies. Basic artificial intelligence technologies generally include sensors, dedicated artificial intelligence chips, cloud computing, distributed storage, big data processing technology, operating / interactive systems, mechatronics and other technologies. Artificial intelligence software technologies mainly include computer vision technology, speech processing technology, natural language processing technology, as well as machine learning / deep learning, autonomous driving, smart transportation and other major directions.

[0075] Intelligent Traffic System (ITS), also known as Intelligent Transportation System, effectively and comprehensively applies advanced science and technology (information technology, computer technology, data communication technology, sensor technology, electronic control technology, automatic control theory, operations research, artificial intelligence, etc.) to transportation, service control and vehicle manufacturing, strengthens the connection between vehicles, roads and users, and thus forms a comprehensive transportation system that ensures safety, improves efficiency, improves the environment and saves energy.

[0076] In order to better understand and illustrate the solutions of the embodiments of the present application, some technical terms involved in the embodiments of the present application are briefly explained below.

[0077] Brushing behavior: Brushing behavior mainly refers to the artificial use of abnormal operating means or techniques to increase the number of views or interactions of a specified work on the Internet, so as to achieve the purpose of increasing popularity or ranking.

[0078] PageRank algorithm: The basic idea of ​​the PageRank algorithm is to define a random walk model on the relationship graph, namely a first-order Markov chain, which describes the behavior of a random walker randomly visiting each node along the relationship graph; under certain conditions, the probability of visiting each node in the extreme case converges to a stable distribution. At this time, the stable probability value of each node is its PageRank value, which indicates the importance of the node.

[0079] Degree centrality: In a network, the greater the degree of a node, the higher the degree centrality of the node, which means that the node is more important in the network.

[0080] Betweenness centrality: An indicator that characterizes the importance of a node by the number of shortest paths passing through the node.

[0081] The solution provided in the embodiments of the present application involves artificial intelligence technology. The technical solution of the present application is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0082] In order to better understand the solution provided by the embodiment of the present application, the solution is described below in conjunction with a specific application scenario.

[0083] In one embodiment, Figure 1 FIG. 1 shows a schematic diagram of the architecture of an identification system applicable to an embodiment of the present application. It can be understood that the identification method provided in the embodiment of the present application can be applied to but not limited to the following applications: Figure 1 In the application scenario shown.

[0084] In this example, Figure 1 As shown, the architecture of the recognition system in this example may include but is not limited to a server 10, a terminal 20 and a database 30. The server 10, the terminal 20 and the database 30 may interact with each other through a network 40.

[0085] The server 10 obtains operation data of multiple users to be identified; based on the operation data of the multiple users to be identified, the server 10 determines the user level of each user to be identified and the user group to which the user belongs through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group; based on the operation data of the multiple users to be identified, the server 10 determines multiple operation behavior segments through segmentation and classification processing, and each operation behavior segment in the multiple operation behavior segments includes multiple operation behaviors arranged in chronological order; the server 10 determines multiple initial targets through evaluation and screening processing based on each user group and each operation behavior segment The operation behavior fragments and the set of users to be identified corresponding to the multiple initial target operation behavior fragments; the server 10 determines a relationship graph based on the multiple initial target operation behavior fragments and the set of users to be identified, wherein the relationship graph includes multiple nodes, and any of the multiple nodes is used to represent any of the multiple initial target operation behavior fragments or any of the users to be identified in the set of users to be identified; based on the relationship graph, the server 10 determines the target operation behavior fragment from the multiple initial target operation behavior fragments through node importance evaluation processing, and determines the users with brushing behavior and the groups with brushing behavior based on the users to be identified corresponding to the target operation behavior fragments. The server 10 sends the list of users with brushing behavior and the list of groups with brushing behavior to the terminal 20, and the server 10 sends the relationship graph to the database 30 for storage.

[0086] It is understandable that the above is only an example and this embodiment is not limited here.

[0087] Among them, terminals include but are not limited to smart phones (such as Android phones, iOS phones, etc.), mobile phone simulators, tablet computers, laptops, digital broadcast receivers, MIDs (Mobile Internet Devices), PDAs (Personal Digital Assistants), intelligent voice interaction devices, smart home appliances, car terminals, etc.

[0088] The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server or server cluster that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), as well as big data and artificial intelligence platforms.

[0089] Cloud computing is a computing model that distributes computing tasks on a resource pool composed of a large number of computers, allowing various application systems to obtain computing power, storage space and information services as needed. The network that provides resources is called a "cloud". From the user's perspective, the resources in the "cloud" are infinitely scalable and can be obtained at any time, used on demand, expanded at any time, and paid for by use.

[0090] As a provider of basic cloud computing capabilities, a cloud computing resource pool (referred to as a cloud platform, generally referred to as an IaaS (Infrastructure as a Service) platform) will be established, and various types of virtual resources will be deployed in the resource pool for external customers to choose to use. The cloud computing resource pool mainly includes: computing devices (virtualized machines, including operating systems), storage devices, and network devices.

[0091] According to the logical function division, the PaaS (Platform as a Service) layer can be deployed on the IaaS (Infrastructure as a Service) layer, and the SaaS (Software as a Service) layer can be deployed on the PaaS layer. SaaS can also be deployed directly on IaaS. PaaS is a platform for software operation, such as databases, web containers, etc. SaaS is a variety of business software, such as web portals, SMS mass senders, etc. Generally speaking, SaaS and PaaS are upper layers relative to IaaS.

[0092] The so-called artificial intelligence cloud service is generally also called AIaaS (AI as a Service). This is the service mode of a mainstream artificial intelligence platform. Specifically, the AIaaS platform will split several common AI services and provide independent or packaged services in the cloud. This service model is similar to opening an AI theme mall: all developers can access and use one or more artificial intelligence services provided by the platform through the API interface. Some senior developers can also use the AI ​​framework and AI infrastructure provided by the platform to deploy and operate their own cloud artificial intelligence services.

[0093] The above network may include but is not limited to: wired network, wireless network, wherein the wired network includes: local area network, metropolitan area network and wide area network, and the wireless network includes: Bluetooth, Wi-Fi and other networks that realize wireless communication. The specific can also be determined based on the actual application scenario requirements and is not limited here.

[0094] See also Figure 2 , Figure 2 The flowchart of an identification method provided by an embodiment of the present application is shown, wherein the method can be executed by any electronic device, such as a server, etc. As an optional implementation, the method can be executed by a server. For the convenience of description, in the description of some optional embodiments below, the server will be used as an example to illustrate the execution subject of the method. Figure 2 As shown, the identification method provided in the embodiment of the present application includes the following steps:

[0095] S201, obtaining operation data of multiple users to be identified.

[0096] Specifically, operation data includes the number of browsing, liking, commenting, forwarding, and collecting operations.

[0097] S202, based on the operation data of the multiple users to be identified, determine the user level of each user to be identified and the user group to which the user belongs through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group.

[0098] Specifically, user levels are, for example, level 1, level 2, level 3, level 4, level 5, level 6, level 7, level 8 and level 9. Level 1, level 2, level 3, level 4, level 5, level 6, level 7, level 8 and level 9 are from low to high. The higher the user level, the more times the user to be identified has performed operations. The more times the user to be identified has performed operations, the greater the probability that the user to be identified is a brushing user. The users to be identified with the same user level are assigned to the same user group, that is, one user group corresponds to one user level, and different user groups correspond to different user levels.

[0099] S203, based on the operation data of the multiple users to be identified, a plurality of operation behavior segments are determined through segmentation and classification processing, wherein each of the plurality of operation behavior segments includes a plurality of operation behaviors arranged in chronological order.

[0100] Specifically, the operation data includes the operation behavior sequence of the user to be identified. For example, after the operation behavior of the user to be identified on the video platform is recorded in the log, the operation behavior sequence of the user to be identified can be generated according to the order in which the user to be identified performs the operations; the operation behavior sequence of the user to be identified is processed by segmentation and classification to obtain multiple operation behavior fragments, one of the multiple operation behavior fragments includes login, search and browse arranged in chronological order, and login, search and browse are all operation behaviors, and another of the multiple operation behavior fragments includes like, comment and search arranged in chronological order, and like, comment and search are all operation behaviors.

[0101] S204, based on each user group and each operation behavior segment, a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments are determined through evaluation and screening processing.

[0102] Specifically, based on user groups corresponding to different user levels and each operation behavior fragment, multiple initial target operation behavior fragments and the user sets to be identified corresponding to the multiple initial target operation behavior fragments are identified; wherein the initial target operation behavior fragment is a suspected brushing operation behavior fragment, and the user set to be identified corresponding to the initial target operation behavior fragment is a suspected brushing user set.

[0103] S205, based on the multiple initial target operation behavior fragments and the set of users to be identified, determine a relationship graph, the relationship graph includes multiple nodes, and any node in the multiple nodes is used to represent any initial target operation behavior fragment in the multiple initial target operation behavior fragments or any user to be identified in the set of users to be identified.

[0104] Specifically, the relationship diagram is such as a relationship map; multiple initial target operation behavior fragments and the users to be identified in the set of users to be identified correspond to nodes in the relationship diagram, a node in the relationship diagram can represent an initial target operation behavior fragment, and another node in the relationship diagram can represent a user to be identified in the set of users to be identified.

[0105] S206, based on the relationship graph, through node importance evaluation processing, determine the target operation behavior segment from multiple initial target operation behavior segments, and based on the to-be-identified users corresponding to the target operation behavior segment, determine the users and groups that have engaged in brushing behavior.

[0106] Specifically, multiple users who engage in inflating traffic all engage in inflating traffic for a certain work, and the work is a inflated traffic work. Multiple users who engage in inflating traffic constitute a group that engages in inflating traffic.

[0107] In an embodiment of the present application, operation data of multiple users to be identified are obtained; based on the operation data of the multiple users to be identified, the user level of each user to be identified among the multiple users to be identified and the user group to which the user belongs are determined through weighted processing, and the users to be identified with the same user level among the multiple users to be identified are assigned to the same user group; based on the operation data of the multiple users to be identified, multiple operation behavior fragments are determined through segmentation and classification processing, and each operation behavior fragment in the multiple operation behavior fragments includes multiple operation behaviors arranged in chronological order; based on each user group and each operation behavior fragment, multiple initial target operation behavior fragments and a set of users to be identified corresponding to the multiple initial target operation behavior fragments are determined through evaluation and screening processing; based on the multiple initial target operation behavior fragments and the set of users to be identified, a relationship graph is determined, and the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or the set of users to be identified. any user to be identified; based on the relationship graph, through node importance evaluation processing, determine the target operation behavior fragment from multiple initial target operation behavior fragments, and based on the user to be identified corresponding to the target operation behavior fragment, determine the users with brushing behavior and the group with brushing behavior; in this way, determine the user level of each user to be identified, the higher the user level, the more times the user to be identified has performed brushing behaviors, and the more times the brushing behaviors are performed, the greater the probability that the user to be identified is a brushing user; based on user groups with different user levels and each operation behavior fragment, identify multiple initial target operation behavior fragments (suspected brushing operation behavior fragments) and the user sets to be identified (suspected brushing user sets) corresponding to the multiple initial target operation behavior fragments; based on the multiple initial target operation behavior fragments and the user set to be identified, construct a relationship graph, and based on the relationship graph, determine the users with brushing behavior and the group with brushing behavior, thereby improving the accuracy of identifying the group with brushing behavior.

[0108] In one embodiment, based on the operation data of the plurality of users to be identified, the user level of each of the plurality of users to be identified is determined by weighted processing, including steps A1-A3:

[0109] Step A1, for each of the multiple users to be identified, based on the operation data of each user to be identified within a preset time period, through classification processing, multiple operation behaviors of each user to be identified and the number of operations corresponding to each of the multiple operation behaviors are determined.

[0110] Specifically, the preset time period is, for example, 1 hour, 6 hours, 12 hours, 1 day, 3 days, etc., and the various operation behaviors are, for example, browsing, liking, commenting, etc. For example, Count (browsing) can be used to represent the number of browsing operations generated by the to-be-identified user within the preset time period, Count (liking) can represent the number of like operations generated by the to-be-identified user within the preset time period, and Count (commenting) can represent the number of comment operations generated by the to-be-identified user within the preset time period.

[0111] Step A2, based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors, determine the operation behavior index corresponding to each to-be-identified user through weighted summation processing.

[0112] Specifically, the difficulty of completing various operation behaviors is different. For example, the difficulty of completing a comment is greater than the difficulty of completing a like, and the difficulty of completing a like is greater than the difficulty of completing a browse. The preset weight corresponding to an operation behavior is used to characterize the difficulty of completing the operation behavior. For example, the preset weight corresponding to a comment is α3, the preset weight corresponding to a like is α2, and the preset weight corresponding to a browse is α1, where α3>α2>α1.

[0113] For example, in the calculation formula (1) of the operation behavior index corresponding to each user to be identified, the number of operations corresponding to each operation behavior among various operation behaviors and the preset weight corresponding to each operation behavior are multiplied to obtain the weighted value corresponding to each operation behavior; the weighted values ​​corresponding to various operation behaviors are summed to determine the operation behavior index corresponding to each user to be identified; formula (1) is as follows:

[0114] Operation behavior index = α1×Count(views)+α2×Count(likes)+α3×Count(comments)+…Formula (1)

[0115] Among them, Count(browsed) represents the number of operations corresponding to browsing generated by the user to be identified within the preset time period, Count(likes) represents the number of operations corresponding to likes generated by the user to be identified within the preset time period, and Count(comments) represents the number of operations corresponding to comments generated by the user to be identified within the preset time period; the preset weight corresponding to comments is α3, the preset weight corresponding to likes is α2, and the preset weight corresponding to browsing is α1; α1×Count(browsed) represents the weighted value corresponding to browsing, α2×Count(likes) represents the weighted value corresponding to likes, and α3×Count(comments) represents the weighted value corresponding to comments.

[0116] Step A3: Based on the operation behavior index corresponding to each user to be identified and a preset classification threshold, a classification process is performed to determine the user level of each user to be identified.

[0117] Specifically, for example, the preset grading thresholds are min, n1, n2, n3, …, max, where min is the minimum value of the operation behavior index, max is the maximum value of the operation behavior index, and min < n1 < n2 < n3 < … < max; if the operation behavior index corresponding to a user to be recognized is between min and n1, the user level of the user to be recognized is grading 1; when the operation behavior index corresponding to a user to be recognized is between n1 and n2, the user level of the user to be recognized is grading 2, and so on, that is, the larger the operation behavior index corresponding to a user to be recognized, the higher the user level of the user to be recognized.

[0118] In one embodiment, based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors, through weighted summation processing, the operation behavior index corresponding to each user to be recognized is determined, including:

[0119] Perform a product process on the number of operations corresponding to each operation behavior and the preset weight corresponding to each operation behavior among various operation behaviors to obtain the weighted value corresponding to each operation behavior;

[0120] Perform a summation process on the weighted values corresponding to various operation behaviors to determine the operation behavior index corresponding to each user to be recognized.

[0121] Specifically, for example, in the calculation formula (1) of the operation behavior index corresponding to each user to be recognized, perform a product process on the number of operations corresponding to each operation behavior and the preset weight corresponding to each operation behavior among various operation behaviors to obtain the weighted value corresponding to each operation behavior; perform a summation process on the weighted values corresponding to various operation behaviors to determine the operation behavior index corresponding to each user to be recognized; the formula (1) is as follows:

[0122] Operation behavior index = α1 × Count(View) + α2 × Count(Like) + α3 × Count(Comment) + … Formula (1)

[0123] Among them, Count(View) represents the number of operations corresponding to the view generated by the user to be recognized within the preset time period, Count(Like) represents the number of operations corresponding to the like generated by the user to be recognized within the preset time period, and Count(Comment) represents the number of operations corresponding to the comment generated by the user to be recognized within the preset time period; the preset weight corresponding to the comment is α3, the preset weight corresponding to the like is α2, and the preset weight corresponding to the view is α1; α1 × Count(View) represents the weighted value corresponding to the view, α2 × Count(Like) represents the weighted value corresponding to the like, and α3 × Count(Comment) represents the weighted value corresponding to the comment.

[0124] In one embodiment, based on the operation data of multiple users to be identified, multiple operation behavior segments are determined through segmentation and classification processing, including steps B1-B3:

[0125] Step B1, for each user to be identified among the multiple users to be identified, based on the operation data of each user to be identified, determining the operation behavior sequence of each user to be identified.

[0126] Specifically, the operation data of a user to be identified includes the operation behavior sequence of the user to be identified. For example, after the operation behavior of a user to be identified on the video platform is recorded in the log, the operation behavior sequence of the user to be identified can be generated according to the order in which the user to be identified performs the operations.

[0127] Step B2, based on the operation behavior sequence of each user to be identified, the preset operation behavior segment length and the preset sliding window length, segmentation processing is performed to determine multiple operation behavior segments of each user to be identified, each of the multiple operation behavior segments including multiple operation behaviors performed by each user to be identified in chronological order.

[0128] Specifically, for example, Figure 3 As shown, for an operation behavior sequence of a user to be identified, the length of the operation behavior segment is L, and the length of the sliding window (window sliding distance) is w. First, the sliding window (window) is used to obtain an operation behavior segment 1 of length L. Operation behavior segment 1 includes login, search, browsing and likes performed in chronological order; then the sliding window is slid a distance w, that is, the window slides w, and the operation behavior segment 2 can be obtained through the sliding window. Operation behavior segment 2 includes likes, searches, comments, etc. performed in chronological order, and so on, until the sliding window slides to the last operation behavior in the operation behavior sequence; in this way, the operation behavior sequence of the user to be identified can be divided into multiple operation behavior segments.

[0129] Step B3: performing classification processing based on the multiple operation behavior fragments of each user to be identified, and determining the multiple operation behavior fragments corresponding to each user to be identified.

[0130] Specifically, for example, if multiple operation behaviors included in an operation behavior segment X are the same as multiple operation behaviors included in an operation behavior segment Y, it is determined that the operation behavior segment X and the operation behavior segment Y are the same operation behavior segment.

[0131] In one embodiment, based on each user group and each operation behavior segment, multiple initial target operation behavior segments and user sets to be identified corresponding to the multiple initial target operation behavior segments are determined through evaluation and screening, including:

[0132] Based on each user group and each operation behavior segment, determining the growth trend value of each operation behavior segment through growth trend evaluation processing;

[0133] Based on the growth trend values ​​of various operation behavior segments, a plurality of initial target operation behavior segments and sets of users to be identified corresponding to the plurality of initial target operation behavior segments are determined through screening processing.

[0134] Specifically, based on user groups corresponding to different user levels and each operation behavior fragment, multiple initial target operation behavior fragments and the user sets to be identified corresponding to the multiple initial target operation behavior fragments are identified; wherein the initial target operation behavior fragment is a suspected brushing operation behavior fragment, and the user set to be identified corresponding to the initial target operation behavior fragment is a suspected brushing user set.

[0135] The higher the user level, the greater the ratio of abnormal operation behavior fragments. By counting the proportion of various operation behavior fragments in user groups corresponding to N user levels, we can identify suspected brushing operation behavior fragments by the change of proportion, where N is a positive integer. For example, for a certain operation behavior fragment, the proportions obtained are: P1, P2, P3, ... P N-1 , P N , P1, P2, P3, ... P N-1 , P N represent the proportion of this type of operation behavior fragment in user group 1, user group 2, user group 3, ..., user group N-1, and user group N respectively. User group 1, user group 2, user group 3, ..., user group N-1, and user group N correspond to level 1, level 2, level 3, ..., level N-1, and level N in the N user levels respectively. The calculation formula (2) for the growth trend value of this type of operation behavior fragment is as follows:

[0136]

[0137] in, is the continuous growth judgment value of the proportion corresponding to this type of operation behavior segment. The continuous growth judgment value of the proportion corresponding to this type of operation behavior segment mainly evaluates whether the growth is continuous growth. Its value range is between 0 and 1. When the value is 0, it means no growth; when the value is 1, it means continuous growth; (P N -P1) is the continuous growth rate of the proportion corresponding to this type of operation behavior segment. The continuous growth rate of the proportion corresponding to this type of operation behavior segment is the absolute value of the proportion growth from the proportion of level 1 to the proportion of level N. Its value range is 0-1. The larger the value, the greater the increase in proportion. It is the continuous growth multiple of the proportion corresponding to this type of operation behavior fragment. The continuous growth multiple of the proportion corresponding to this type of operation behavior fragment mainly evaluates the ratio of the proportion of level N to the sum of the proportion of level 1 and level N. Its value range is 0-1. The larger the value, the greater the multiple of level N relative to level 1.

[0138] In one embodiment, based on each user group and each operation behavior segment, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing, including steps C1-C2:

[0139] Step C1, based on each user group and each operation behavior segment, determine the proportion of each operation behavior segment in each user group through proportion calculation processing.

[0140] Specifically, for example, for a certain operation behavior segment, the proportions obtained are: P1, P2, P3, ... P N-1 , P N , P1, P2, P3, ... P N-1 , P N They respectively represent the proportion of this type of operation behavior fragment in user group 1, user group 2, user group 3, ..., user group N-1, and user group N. User group 1, user group 2, user group 3, ..., user group N-1, and user group N correspond to level 1, level 2, level 3, ..., level N-1, and level N in the N user levels respectively.

[0141] For example, Figure 4 As shown in the figure, for a certain operation behavior fragment, the proportions obtained are: 0.30%, 0.35%, 0.80%, 0.81%, 1.38%, 2.67%, 5.67%, 8.46% and 9.76% respectively; 0.30%, 0.35%, 0.80%, 0.81%, 1.38%, 2.67%, 5.67%, 8.46% and 9.76% represent the operation behavior fragment in user group 1, user group 2, user group 3, user group 4 and user group 5 respectively. , user group 1, user group 2, user group 3, user group 4, user group 5, user group 6, user group 7, user group 8 and user group 9 correspond to level 1, level 2, level 3, level 4, level 5, level 6, level 7, level 8 and level 9 in 9 user levels respectively; level 1, level 2, level 3, level 4, level 5, level 6, level 7, level 8 and level 9 from low to high.

[0142] Step C2: Based on the proportion of each operation behavior segment in each user group, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing.

[0143] Specifically, for example, the calculation formula (2) for the growth trend value of a certain operation behavior segment is as follows:

[0144]

[0145] in, is the continuous growth judgment value of the proportion corresponding to this type of operation behavior segment. The continuous growth judgment value of the proportion corresponding to this type of operation behavior segment mainly evaluates whether the growth is continuous growth. Its value range is between 0 and 1. When the value is 0, it means no growth; when the value is 1, it means continuous growth; (P N -P1) is the continuous growth rate of the proportion corresponding to this type of operation behavior segment. The continuous growth rate of the proportion corresponding to this type of operation behavior segment is the absolute value of the proportion growth from the proportion of level 1 to the proportion of level N. Its value range is 0-1. The larger the value, the greater the increase in proportion. It is the continuous growth multiple of the proportion corresponding to this type of operation behavior fragment. The continuous growth multiple of the proportion corresponding to this type of operation behavior fragment mainly evaluates the ratio of the proportion of level N to the sum of the proportion of level 1 and level N. Its value range is 0-1. The larger the value, the greater the multiple of level N relative to level 1.

[0146] In one embodiment, based on the proportion of each operation behavior segment in each user group, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing, including:

[0147] Based on the proportion of each operation behavior segment in each user group, through proportion calculation processing, determine the proportion continuous growth judgment value corresponding to each operation behavior segment, the proportion continuous growth amplitude corresponding to each operation behavior segment and the proportion continuous growth multiple corresponding to each operation behavior segment. The proportion continuous growth judgment value is used to indicate whether the proportion of each operation behavior segment in each user group is proportional to the user level corresponding to each user group.

[0148] Based on the judgment value of continuous growth in proportion, the amplitude of continuous growth in proportion and the multiple of continuous growth in proportion, the growth trend value of each operation behavior segment is determined through multiplication processing.

[0149] Specifically, for example, the calculation formula (2) for the growth trend value of a certain operation behavior segment is as follows:

[0150]

[0151] in, is the continuous growth judgment value of the proportion corresponding to this type of operation behavior segment. The continuous growth judgment value of the proportion corresponding to this type of operation behavior segment mainly evaluates whether the growth is continuous growth. Its value range is between 0 and 1. When the value is 0, it means no growth; when the value is 1, it means continuous growth; (P N -P1) is the continuous growth rate of the proportion corresponding to this type of operation behavior segment. The continuous growth rate of the proportion corresponding to this type of operation behavior segment is the absolute value of the proportion growth from the proportion of level 1 to the proportion of level N. Its value range is 0-1. The larger the value, the greater the increase in proportion. It is the continuous growth multiple of the proportion corresponding to this type of operation behavior fragment. The continuous growth multiple of the proportion corresponding to this type of operation behavior fragment mainly evaluates the ratio of the proportion of level N to the sum of the proportion of level 1 and level N. Its value range is 0-1. The larger the value, the greater the multiple of level N relative to level 1.

[0152] In one embodiment, based on the growth trend values ​​of various operation behavior segments, multiple initial target operation behavior segments and user sets to be identified corresponding to the multiple initial target operation behavior segments are determined through screening processing, including:

[0153] If the growth trend value of any one of the various operation behavior segments is greater than a preset growth trend threshold, then determining any one of the operation behavior segments as an initial target operation behavior segment;

[0154] The to-be-identified users corresponding to various initial target operation behavior fragments are constructed into a to-be-identified user set corresponding to various initial target operation behavior fragments.

[0155] Specifically, for example, the preset growth trend threshold is t, and the growth trend value of a certain operation behavior segment is greater than t, which means that the growth trend of this operation behavior segment is more obvious as the user level increases. This operation behavior segment is determined as the initial target operation behavior segment, and the initial target operation behavior segment is a suspected brushing operation behavior segment. The set of users to be identified corresponding to the initial target operation behavior segment is a set of suspected brushing users.

[0156] In one embodiment, based on the relationship graph, a target operation behavior segment is determined from a plurality of initial target operation behavior segments through node importance evaluation processing, including:

[0157] Based on the relationship graph, the node importance corresponding to any one of the multiple initial target operation behavior segments is determined through node importance evaluation processing, and the node importance corresponding to any one of the initial target operation behavior segments is the stable probability value of the node in the relationship graph corresponding to any one of the initial target operation behavior segments;

[0158] If the node importance corresponding to any initial target operation behavior segment is greater than a preset importance threshold, then any initial target operation behavior segment is determined as a target operation behavior segment.

[0159] Specifically, for example, Figure 5 The relationship diagram shown includes 12 nodes, which are o_p1, o_p2, o_p3, o_p4, o_p5, o_p6, o_p9, u1, u2, u3, u4 and u5. o_p1, o_p2, o_p3, o_p4, o_p5, o_p6 and o_p9 represent 7 initial target operation behavior fragments, and u1, u2, u3, u4 and u5 represent 7 users to be identified in the set of users to be identified. The weight of the edge between a node (an initial target operation behavior fragment) and another node (a user to be identified) can be the growth trend value of the initial target operation behavior fragment multiplied by the number of times the user to be identified executes the operation behavior fragment. For example, the growth trend value of o_p2 is 0.6, the number of times u1 executes o_p2 is 8, and the weight of the edge between o_p2 and u1 is 0.6×8=4.8.

[0160] Node importance evaluation processes such as PageRank algorithm, degree centrality evaluation, betweenness centrality evaluation, etc.

[0161] Based on the relationship graph, the node importance corresponding to any of the multiple initial target operation behavior fragments is determined by the PageRank algorithm. The node importance corresponding to any of the initial target operation behavior fragments is shown in formula (3):

[0162]

[0163] Among them, PR (node n ) i+1 It indicates the PageRank value of the nth node in the relationship graph in the i+1th round. The PageRank value indicates the importance of the node. j ) i represents the PageRank value of the jth node in the i-th round in the relationship graph, L(node j ) represents the degree of the j-th node. The degree of the j-th node is the number of nodes connected to the j-th node. n represents the n-th node. N represents the total number of nodes in the relationship graph. n ranges from 1 to N.

[0164] For example, Figure 5 The relationship diagram shown includes 12 nodes, that is, N=12, the jth node and the nth node are connected to each other, and j is not equal to n; the first round (i=1) is the initial value, and the initial value is: the PageRank value of each node in the N nodes is 1 / N, Figure 5 The PageRank values ​​of the five nodes o_p2, u1, u2, u3, and u4 are all 1 / 12. o_p2 is connected to u1, u2, u3, and u4 respectively. Formula (3) calculates the PageRank value of o_p2 through u1, u2, u3, and u4.

[0165] It should be noted that if the PageRank value of o_p2 obtained after the i-th round is the same as the PageRank value obtained after the i+1-th round, it means that the PageRank value has converged, that is, the node importance of o_p2 is determined to be the PageRank value obtained after the i+1-th round.

[0166] In one embodiment, for example, Figure 5 In the relationship diagram shown in formula (3), the nth node is o_p2, and the jth nodes are u1, u2, u3 and u4 respectively; if the jth node is u1, and u1 connects two nodes (o_p1 and o_p2), then 0.6×8 / (0.3×32+0.6×8) is used as the weight corresponding to u1, and the weight corresponding to u1 is 1 / L(node j ), that is, the L(node j ), where 0.3×32 represents the weight of the edge between u1 and o_p1, and 0.6×8 represents the weight of the edge between u1 and o_p2; u2, u3, and u4 correspond to L(node j ) and so on.

[0167] In one embodiment, based on the to-be-identified user corresponding to the target operation behavior segment, determining the user who has engaged in the brushing behavior and the group who has engaged in the brushing behavior includes:

[0168] For any user to be identified corresponding to the target operation behavior segment, if the number of times any user to be identified performs the target operation behavior segment is greater than a preset number threshold, then any user to be identified is determined as a user who has engaged in brushing behavior;

[0169] Based on the users who have engaged in inflating traffic, the group that has engaged in inflating traffic is determined through statistical estimation.

[0170] Specifically, for example, a group of people who engage in brushing behavior generally focus on brushing the volume of the same batch of works. For a group of users who engage in brushing behavior, if the works brushed by this group of users are consistent within the same period of time, then this group of users is determined to be a group that engages in brushing behavior.

[0171] In one embodiment, based on each user who has engaged in the inflating behavior, a group that has engaged in the inflating behavior is determined through statistical estimation processing, including:

[0172] If it is determined that multiple users who have engaged in brushing behavior have engaged in brushing behavior on the same work, the same work will be determined as a brushed work, and the multiple users who have engaged in brushing behavior will be determined as a group that has engaged in brushing behavior.

[0173] Specifically, after determining multiple users who engage in brushing behavior, extract the works operated by these multiple users who engage in brushing behavior. Works operated by multiple users who engage in brushing behavior can be determined as brushing works. Groups that engage in brushing behavior generally focus on brushing the same batch of works. For a group of users who engage in brushing behavior, if the works brushed by this group of users at the same time are consistent, then this group of users is determined to be a group that engages in brushing behavior. Operations such as blocking the accounts of groups that engage in brushing behavior, and limiting the flow of brushing works.

[0174] The application of the embodiments of the present application has at least the following beneficial effects:

[0175] Determine the user level of each user to be identified. The higher the user level, the more times the user to be identified has performed brushing behaviors. The more times the brushing behaviors are performed, the greater the probability that the user to be identified is a brushing user. Based on user groups with different user levels and each operation behavior fragment, identify multiple initial target operation behavior fragments (suspected brushing operation behavior fragments) and sets of users to be identified corresponding to the multiple initial target operation behavior fragments (suspected brushing user sets). Based on the multiple initial target operation behavior fragments and the sets of users to be identified, construct a relationship graph, and based on the relationship graph, determine users with brushing behaviors and groups with brushing behaviors, thereby improving the accuracy of identifying groups with brushing behaviors.

[0176] In order to better understand the method provided in the embodiment of the present application, the scheme of the embodiment of the present application is further explained below with reference to examples of specific application scenarios.

[0177] In a specific application scenario, such as a brushing behavior scenario, see Figure 6 , shows a processing flow of an identification method, such as Figure 6 As shown, the processing flow of the identification method provided in the embodiment of the present application includes the following steps:

[0178] S601: The server obtains operation data of multiple users to be identified.

[0179] Specifically, for example, Figure 7The whole process from the user's need to increase the number of views to the end of the video's increase in views is shown, which specifically includes: the user publishes the video; the user has a need to increase the number of views; the user purchases the increase in views service on some increase in views platforms; the user pays and submits the video link; a group performs batch operations, such as likes and comments; the increase in views is completed, and the popularity of the video increases. The server can obtain the operation data of multiple users to be identified from the batch operations of a group or multiple groups.

[0180] S602, the server determines, for each of the multiple users to be identified, multiple operation behaviors of each user to be identified and the number of operations corresponding to each of the multiple operation behaviors through classification processing based on the operation data of each user to be identified within a preset time period.

[0181] S603: The server determines the operation behavior index corresponding to each to-be-identified user through weighted summation based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors.

[0182] S604: The server performs classification processing based on the operation behavior index corresponding to each user to be identified and a preset classification threshold, and determines the user level of each user to be identified and the user group corresponding to each user level.

[0183] S605, the server performs segmentation and classification processing based on the operation behavior sequence of each user to be identified, the preset operation behavior segment length and the preset sliding window length, and determines multiple operation behavior segments of each user to be identified.

[0184] Specifically, the operation data of a user to be identified includes an operation behavior sequence of the user to be identified.

[0185] S606: The server determines the growth trend value of each operation behavior segment through growth trend evaluation processing based on the user groups corresponding to each user level and each operation behavior segment.

[0186] S607: The server determines a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through screening based on the growth trend values ​​of the various operation behavior segments.

[0187] S608: The server determines a relationship graph based on the multiple initial target operation behavior fragments and the set of users to be identified.

[0188] Specifically, for example, Figure 5The relationship diagram shown includes 12 nodes, which are o_p1, o_p2, o_p3, o_p4, o_p5, o_p6, o_p9, u1, u2, u3, u4 and u5. o_p1, o_p2, o_p3, o_p4, o_p5, o_p6 and o_p9 represent 7 initial target operation behavior fragments, and u1, u2, u3, u4 and u5 represent 7 users to be identified in the set of users to be identified. The weight of the edge between a node (an initial target operation behavior fragment) and another node (a user to be identified) can be the growth trend value of the initial target operation behavior fragment multiplied by the number of times the user to be identified executes the operation behavior fragment. For example, the growth trend value of o_p2 is 0.6, the number of times u1 executes o_p2 is 8, and the weight of the edge between o_p2 and u1 is 0.6×8=4.8.

[0189] S609, based on the relationship graph, the server determines the node importance corresponding to any one of the multiple initial target operation behavior fragments through node importance evaluation processing; if the node importance corresponding to any one of the initial target operation behavior fragments is greater than a preset importance threshold, then any one of the initial target operation behavior fragments is determined as the target operation behavior fragment.

[0190] S610, the server determines the users who have engaged in brushing behavior, the groups who have engaged in brushing behavior, and the corresponding brushing works based on the to-be-identified users corresponding to the target operation behavior segment.

[0191] The application of the embodiments of the present application has at least the following beneficial effects:

[0192] Based on the prior knowledge that brushing users (users who engage in brushing behaviors) perform more operations, different classifications (user levels) are obtained. The higher the classification, the greater the proportion of brushing users. In this way, the brushing users are initially recalled. Based on the occurrence ratio of operation behavior fragments in different classifications, operation behavior fragments with abnormal changes in occurrence ratios are extracted as abnormal operation behavior fragments (target operation behavior fragments). Based on the abnormal operation behavior fragments (target operation behavior fragments), brushing users, brushing works and brushing groups (groups that engage in brushing behaviors) are determined. In this way, the prevention of brushing behaviors is achieved from points (brushing works), lines (brushing users) to surfaces (brushing groups).

[0193] The present application also provides an identification device. The structural diagram of the identification device is as follows: Figure 8 As shown, the identification device 80 includes a first processing module 801 , a second processing module 802 , a third processing module 803 , a fourth processing module 804 , a fifth processing module 805 and a sixth processing module 806 .

[0194] The first processing module 801 is used to obtain operation data of multiple users to be identified;

[0195] The second processing module 802 is used to determine the user level of each of the multiple users to be identified and the user group to which the user belongs through weighted processing based on the operation data of the multiple users to be identified, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group;

[0196] The third processing module 803 is used to determine multiple operation behavior segments based on the operation data of multiple users to be identified through segmentation and classification processing, where each of the multiple operation behavior segments includes multiple operation behaviors arranged in chronological order;

[0197] The fourth processing module 804 is used to determine a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through evaluation and screening processing based on each user group and each operation behavior segment;

[0198] A fifth processing module 805 is used to determine a relationship graph based on the multiple initial target operation behavior segments and the set of users to be identified, wherein the relationship graph includes multiple nodes, and any node in the multiple nodes is used to represent any initial target operation behavior segment in the multiple initial target operation behavior segments or any user to be identified in the set of users to be identified;

[0199] The sixth processing module 806 is used to determine the target operation behavior segment from multiple initial target operation behavior segments through node importance evaluation processing based on the relationship graph, and determine the users and groups that have engaged in brushing behavior based on the to-be-identified users corresponding to the target operation behavior segment.

[0200] In one embodiment, the second processing module 802 is specifically configured to:

[0201] For each of the multiple users to be identified, based on the operation data of each user to be identified within a preset time period, determining multiple operation behaviors of each user to be identified and the number of operations corresponding to each of the multiple operation behaviors through classification processing;

[0202] Based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors, an operation behavior index corresponding to each to-be-identified user is determined through weighted summation processing;

[0203] Based on the operation behavior index corresponding to each user to be identified and the preset classification threshold, classification processing is performed to determine the user level of each user to be identified.

[0204] In one embodiment, the second processing module 802 is specifically configured to:

[0205] The number of operations corresponding to each operation behavior in various operation behaviors and the preset weight corresponding to each operation behavior are multiplied to obtain a weighted value corresponding to each operation behavior;

[0206] The weighted values ​​corresponding to various operation behaviors are summed up to determine the operation behavior index corresponding to each user to be identified.

[0207] In one embodiment, the third processing module 803 is specifically configured to:

[0208] For each of the multiple users to be identified, based on the operation data of each of the users to be identified, determining an operation behavior sequence of each of the users to be identified;

[0209] Based on the operation behavior sequence of each to-be-identified user, the preset operation behavior segment length, and the preset sliding window length, segmentation processing is performed to determine multiple operation behavior segments of each to-be-identified user, each of the multiple operation behavior segments including multiple operation behaviors performed by each to-be-identified user in chronological order;

[0210] Based on the multiple operation behavior fragments of each to-be-identified user, classification processing is performed to determine the multiple operation behavior fragments corresponding to each to-be-identified user.

[0211] In one embodiment, the fourth processing module 804 is specifically configured to:

[0212] Based on each user group and each operation behavior segment, determining the growth trend value of each operation behavior segment through growth trend evaluation processing;

[0213] Based on the growth trend values ​​of various operation behavior segments, a plurality of initial target operation behavior segments and sets of users to be identified corresponding to the plurality of initial target operation behavior segments are determined through screening processing.

[0214] In one embodiment, the fourth processing module 804 is specifically configured to:

[0215] Based on each user group and each operation behavior segment, determine the proportion of each operation behavior segment in each user group through ratio calculation processing;

[0216] Based on the proportion of each operation behavior segment in each user group, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing.

[0217] In one embodiment, the fourth processing module 804 is specifically configured to:

[0218] Based on the proportion of each operation behavior segment in each user group, through proportion calculation processing, determine the proportion continuous growth judgment value corresponding to each operation behavior segment, the proportion continuous growth amplitude corresponding to each operation behavior segment and the proportion continuous growth multiple corresponding to each operation behavior segment. The proportion continuous growth judgment value is used to indicate whether the proportion of each operation behavior segment in each user group is proportional to the user level corresponding to each user group.

[0219] Based on the judgment value of continuous growth in proportion, the amplitude of continuous growth in proportion and the multiple of continuous growth in proportion, the growth trend value of each operation behavior segment is determined through multiplication processing.

[0220] In one embodiment, the fourth processing module 804 is specifically configured to:

[0221] If the growth trend value of any one of the various operation behavior segments is greater than a preset growth trend threshold, then determining any one of the operation behavior segments as an initial target operation behavior segment;

[0222] The to-be-identified users corresponding to various initial target operation behavior fragments are constructed into a to-be-identified user set corresponding to various initial target operation behavior fragments.

[0223] In one embodiment, the sixth processing module 806 is specifically configured to:

[0224] Based on the relationship graph, the node importance corresponding to any one of the multiple initial target operation behavior segments is determined through node importance evaluation processing, and the node importance corresponding to any one of the initial target operation behavior segments is the stable probability value of the node in the relationship graph corresponding to any one of the initial target operation behavior segments;

[0225] If the node importance corresponding to any initial target operation behavior segment is greater than a preset importance threshold, then any initial target operation behavior segment is determined as a target operation behavior segment.

[0226] In one embodiment, the sixth processing module 806 is specifically configured to:

[0227] For any user to be identified corresponding to the target operation behavior segment, if the number of times any user to be identified performs the target operation behavior segment is greater than a preset number threshold, then any user to be identified is determined as a user who has engaged in brushing behavior;

[0228] Based on the users who have engaged in inflating traffic, the group that has engaged in inflating traffic is determined through statistical estimation.

[0229] In one embodiment, the sixth processing module 806 is specifically configured to:

[0230] If it is determined that multiple users who have engaged in brushing behavior have engaged in brushing behavior on the same work, the same work will be determined as a brushed work, and the multiple users who have engaged in brushing behavior will be determined as a group that has engaged in brushing behavior.

[0231] The application of the embodiments of the present application has at least the following beneficial effects:

[0232] Acquire operation data of multiple users to be identified; determine the user level of each user to be identified and the user group to which the user belongs, based on the operation data of the multiple users to be identified, through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group; determine multiple operation behavior fragments based on the operation data of the multiple users to be identified, through segmentation and classification processing, each of the multiple operation behavior fragments includes multiple operation behaviors arranged in chronological order; determine multiple initial target operation behavior fragments and the user sets to be identified corresponding to the multiple initial target operation behavior fragments through evaluation and screening processing based on each user group and each operation behavior fragment; determine a relationship graph based on the multiple initial target operation behavior fragments and the user set to be identified, the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or any one of the user sets to be identified. Identify users; based on the relationship graph, determine the target operation behavior fragment from multiple initial target operation behavior fragments through node importance evaluation processing, and based on the users to be identified corresponding to the target operation behavior fragment, determine the users who have brushed the volume and the groups who have brushed the volume; in this way, determine the user level of each user to be identified, the higher the user level, the more times the user to be identified has performed brushing behaviors, and the more times the brushing behaviors are performed, the greater the probability that the user to be identified is a brushing user; based on user groups with different user levels and each operation behavior fragment, identify multiple initial target operation behavior fragments (suspected brushing operation behavior fragments) and sets of users to be identified (suspected brushing user sets) corresponding to the multiple initial target operation behavior fragments; based on the multiple initial target operation behavior fragments and the set of users to be identified, construct a relationship graph, and based on the relationship graph, determine the users who have brushed the volume and the groups who have brushed the volume, thereby improving the accuracy of identifying the groups who have brushed the volume.

[0233] The present application also provides an electronic device. The structural diagram of the electronic device is as follows: Fig. 9 As shown, Fig. 9The electronic device 4000 shown includes: a processor 4001 and a memory 4003. The processor 4001 and the memory 4003 are connected, such as through a bus 4002. Optionally, the electronic device 4000 may also include a transceiver 4004, which may be used for data interaction between the electronic device and other electronic devices, such as data transmission and / or data reception. It should be noted that in actual applications, the transceiver 4004 is not limited to one, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present application.

[0234] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array) or other programmable logic devices, transistor logic devices, hardware components or any combination thereof. It may implement or execute various exemplary logic blocks, modules and circuits described in conjunction with the disclosure of this application. Processor 4001 may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a DSP and a microprocessor, etc.

[0235] The bus 4002 may include a path to transmit information between the above components. The bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus. The bus 4002 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig. 9 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0236] The memory 4003 may be a ROM (Read Only Memory) or other types of static storage devices that can store static information and instructions, a RAM (Random Access Memory) or other types of dynamic storage devices that can store information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), a CD-ROM (Compact Disc Read Only Memory) or other optical disk storage, optical disk storage (including compressed optical disk, laser disk, optical disk, digital versatile disk, Blu-ray disk, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium that can be used to carry or store computer programs and can be read by a computer, without limitation herein.

[0237] The memory 4003 is used to store the computer program for executing the embodiment of the present application, and the execution is controlled by the processor 4001. The processor 4001 is used to execute the computer program stored in the memory 4003 to implement the steps shown in the above method embodiment.

[0238] Among them, electronic equipment includes but is not limited to: servers, etc.

[0239] The application of the embodiments of the present application has at least the following beneficial effects:

[0240] Acquire operation data of multiple users to be identified; determine the user level of each user to be identified and the user group to which the user belongs, based on the operation data of the multiple users to be identified, through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group; determine multiple operation behavior fragments based on the operation data of the multiple users to be identified, through segmentation and classification processing, each of the multiple operation behavior fragments includes multiple operation behaviors arranged in chronological order; determine multiple initial target operation behavior fragments and the user sets to be identified corresponding to the multiple initial target operation behavior fragments through evaluation and screening processing based on each user group and each operation behavior fragment; determine a relationship graph based on the multiple initial target operation behavior fragments and the user set to be identified, the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or any one of the user sets to be identified. Identify users; based on the relationship graph, determine the target operation behavior fragment from multiple initial target operation behavior fragments through node importance evaluation processing, and based on the users to be identified corresponding to the target operation behavior fragment, determine the users who have brushed the volume and the groups who have brushed the volume; in this way, determine the user level of each user to be identified, the higher the user level, the more times the user to be identified has performed brushing behaviors, and the more times the brushing behaviors are performed, the greater the probability that the user to be identified is a brushing user; based on user groups with different user levels and each operation behavior fragment, identify multiple initial target operation behavior fragments (suspected brushing operation behavior fragments) and sets of users to be identified (suspected brushing user sets) corresponding to the multiple initial target operation behavior fragments; based on the multiple initial target operation behavior fragments and the set of users to be identified, construct a relationship graph, and based on the relationship graph, determine the users who have brushed the volume and the groups who have brushed the volume, thereby improving the accuracy of identifying the groups who have brushed the volume.

[0241] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps and corresponding contents of the aforementioned method embodiment can be implemented.

[0242] The embodiment of the present application also provides a computer program product, including a computer program, which can implement the steps and corresponding contents of the aforementioned method embodiment when executed by a processor.

[0243] Based on the same principle as the method provided in the embodiment of the present application, the embodiment of the present application further provides a computer program product or a computer program, which includes a computer instruction, and the computer instruction is stored in a computer-readable storage medium. The processor of the computer device reads the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the method provided in any optional embodiment of the present application.

[0244] It should be understood that, although each operation step is indicated by arrows in the flowchart of the embodiment of the present application, the implementation order of these steps is not limited to the order indicated by the arrows. Unless clearly stated herein, in some implementation scenarios of the embodiment of the present application, the implementation steps in each flowchart can be performed in other orders according to demand. In addition, some or all of the steps in each flowchart may include multiple sub-steps or multiple stages based on actual implementation scenarios. Some or all of these sub-steps or stages may be executed at the same time, and each sub-step or stage in these sub-steps or stages may also be executed at different times respectively. In different scenarios of execution time, the execution order of these sub-steps or stages may be flexibly configured according to demand, and the embodiment of the present application does not limit this.

[0245] The above is only an optional implementation method for some implementation scenarios of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the technical concept of the solution of the present application, other similar implementation methods based on the technical ideas of the present application are also within the protection scope of the embodiments of the present application.

Claims

1. A recognition method, characterized in that: include: Obtaining operation data of multiple users to be identified; Based on the operation data of the multiple users to be identified, determining the user level of each user to be identified and the user group to which the user belongs through weighted processing, and the users to be identified with the same user level among the multiple users to be identified belong to the same user group; Based on the operation data of the multiple users to be identified, a plurality of operation behavior segments are determined by segmentation and classification, wherein each of the plurality of operation behavior segments includes a plurality of operation behaviors arranged in chronological order; Based on each user group and each operation behavior segment, determining multiple initial target operation behavior segments and sets of users to be identified corresponding to the multiple initial target operation behavior segments through evaluation and screening processing; Based on the multiple initial target operation behavior fragments and the set of users to be identified, determining a relationship graph, wherein the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior fragments or any one of the users to be identified in the set of users to be identified; Based on the relationship graph, through node importance evaluation processing, the target operation behavior fragment is determined from the multiple initial target operation behavior fragments, and based on the to-be-identified users corresponding to the target operation behavior fragments, the users and groups that have engaged in brushing behavior are determined.

2. The method according to claim 1, characterized in that: The determining, based on the operation data of the plurality of users to be identified, the user level of each of the plurality of users to be identified through weighted processing, comprises: For each of the multiple users to be identified, based on the operation data of each of the users to be identified within a preset time period, determining multiple operation behaviors of each of the users to be identified and the number of operations corresponding to each of the multiple operation behaviors through classification processing; Based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to the various operation behaviors, determining the operation behavior index corresponding to each to-be-identified user through weighted summation processing; Based on the operation behavior index corresponding to each user to be identified and a preset classification threshold, classification processing is performed to determine the user level of each user to be identified.

3. The method according to claim 2, characterized in that The determining of the operation behavior index corresponding to each to-be-identified user through weighted summation based on the number of operations corresponding to various operation behaviors and the preset weights corresponding to various operation behaviors includes: The number of operations corresponding to each operation behavior among various operation behaviors is multiplied by a preset weight corresponding to each operation behavior to obtain a weighted value corresponding to each operation behavior; The weighted values ​​corresponding to the various operation behaviors are summed up to determine the operation behavior index corresponding to each user to be identified.

4. The method according to claim 1, characterized in that: The operation data of the multiple users to be identified are divided and classified to determine multiple operation behavior segments, including: For each to-be-identified user among the multiple to-be-identified users, determining an operation behavior sequence of each to-be-identified user based on the operation data of each to-be-identified user; Based on the operation behavior sequence of each to-be-identified user, a preset operation behavior segment length, and a preset sliding window length, segmentation processing is performed to determine a plurality of operation behavior segments of each to-be-identified user, each of the plurality of operation behavior segments including a plurality of operation behaviors performed by each to-be-identified user in chronological order; Based on the multiple operation behavior fragments of each to-be-identified user, classification processing is performed to determine the multiple operation behavior fragments corresponding to each to-be-identified user.

5. The method according to claim 1, characterized in that: The step of determining a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through evaluation and screening based on each user group and each operation behavior segment includes: Based on each user group and each operation behavior segment, determining a growth trend value of each operation behavior segment through growth trend evaluation processing; Based on the growth trend values ​​of various operation behavior segments, a plurality of initial target operation behavior segments and sets of users to be identified corresponding to the plurality of initial target operation behavior segments are determined through screening processing.

6. The method according to claim 5, characterized in that The determining the growth trend value of each operation behavior segment through growth trend evaluation processing based on each user group and each operation behavior segment includes: Based on each user group and each operation behavior segment, determine the proportion of each operation behavior segment in each user group in each user group through proportion calculation processing; Based on the proportion of each operation behavior segment in each of the user groups, a growth trend value of each operation behavior segment is determined through growth trend evaluation processing.

7. The method according to claim 6, characterized in that The determining the growth trend value of each operation behavior segment through growth trend evaluation processing based on the proportion of each operation behavior segment in each user group in each user group includes: Based on the proportion of each operation behavior segment in each of the user groups, determine a proportion continuous growth judgment value corresponding to each operation behavior segment, a proportion continuous growth amplitude corresponding to each operation behavior segment, and a proportion continuous growth multiple corresponding to each operation behavior segment through proportion calculation processing, wherein the proportion continuous growth judgment value is used to indicate whether the proportion of each operation behavior segment in each user group is proportional to the user level corresponding to each user group; Based on the continuous growth judgment value of the proportion, the continuous growth amplitude of the proportion and the continuous growth multiple of the proportion, the growth trend value of each operation behavior segment is determined through multiplication processing.

8. The method according to claim 5, characterized in that The method of determining a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through screening based on the growth trend values ​​of the various operation behavior segments includes: If the growth trend value of any one of the various operation behavior segments is greater than a preset growth trend threshold, determining the any one of the operation behavior segments as the initial target operation behavior segment; The to-be-identified users corresponding to various initial target operation behavior fragments are constructed into a to-be-identified user set corresponding to various initial target operation behavior fragments.

9. The method according to claim 1, characterized in that: The determining of the target operation behavior fragment from the multiple initial target operation behavior fragments through node importance evaluation based on the relationship graph includes: Based on the relationship graph, determining the node importance corresponding to any one of the multiple initial target operation behavior segments through node importance evaluation processing, wherein the node importance corresponding to any one of the initial target operation behavior segments is the steady probability value of the node in the relationship graph corresponding to any one of the initial target operation behavior segments; If the node importance corresponding to any one of the initial target operation behavior segments is greater than a preset importance threshold, then the any one of the initial target operation behavior segments is determined as the target operation behavior segment.

10. The method according to claim 1, characterized in that The determining of users and groups that have engaged in brushing behavior based on the to-be-identified users corresponding to the target operation behavior fragments includes: For any user to be identified corresponding to the target operation behavior segment, if the number of times that the user to be identified performs the target operation behavior segment is greater than a preset number threshold, the user to be identified is determined as a user who has engaged in brushing behavior; Based on the users who have engaged in inflating traffic, the group that has engaged in inflating traffic is determined through statistical estimation.

11. The method according to claim 10, characterized in that The group of users who engage in volume-boosting behavior is determined through statistical estimation based on each user who engages in volume-boosting behavior, including: If it is determined that multiple users among the users who have engaged in brushing behavior have engaged in brushing behavior on the same work, the same work will be determined as a brushed work, and the multiple users who have engaged in brushing behavior will be determined as a group that has engaged in brushing behavior.

12. An identification device, characterized in that: include: A first processing module is used to obtain operation data of multiple users to be identified; A second processing module is used to determine the user level of each user to be identified and the user group to which each user belongs based on the operation data of the multiple users to be identified through weighted processing, and the users to be identified with the same user level belong to the same user group. A third processing module is used to determine a plurality of operation behavior segments based on the operation data of the plurality of users to be identified by segmentation and classification processing, wherein each of the plurality of operation behavior segments includes a plurality of operation behaviors arranged in chronological order; A fourth processing module is used to determine, based on each user group and each operation behavior segment, a plurality of initial target operation behavior segments and a set of users to be identified corresponding to the plurality of initial target operation behavior segments through evaluation and screening processing; a fifth processing module, configured to determine a relationship graph based on the multiple initial target operation behavior segments and the set of users to be identified, wherein the relationship graph includes multiple nodes, and any one of the multiple nodes is used to represent any one of the multiple initial target operation behavior segments or any one of the users to be identified in the set of users to be identified; The sixth processing module is used to determine the target operation behavior fragment from the multiple initial target operation behavior fragments based on the relationship graph through node importance evaluation processing, and determine the users and groups that have engaged in brushing behavior based on the users to be identified corresponding to the target operation behavior fragments.

13. An electronic device comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to implement the steps of the method according to any one of claims 1 to 11.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 11 are implemented.

Images