Method for dividing user classification and related devices

Abstract

Embodiments of the present disclosure provide a method and apparatus for classifying users, a computer-readable storage medium, and an electronic device. The method includes: obtaining a plurality of user behavior information within a first time period; obtaining, according to the plurality of user behavior information, the time, the number of behaviors, and the information on the total resources used for the behavior of the most recent behavior of the first category for each user; obtaining, according to the time, the number of behaviors, and the information on the total resources used for the behavior of the most recent behavior of the first category for each user, the time threshold, the behavior number threshold, and the information threshold on the total resources used for the behavior of the most recent behavior of the first category for the plurality of users; and identifying, among the plurality of users, the users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior number threshold, and the information on the total resources used for the behavior is greater than or equal to the information threshold on the total resources used for the behavior as the first-level users of the first category. The method of the present disclosure can achieve the classification of users.

Description

Technical Field

[0001] The present disclosure relates to the fields of computer and communication technologies, and in particular, to a method and apparatus for classifying users, a computer-readable storage medium, and an electronic device. Background Art

[0002] Loyal users refer to users who have developed a preference for a specific product or service and thus reuse it. The strategic significance of loyal users has been recognized by many enterprises. Currently, the classification of "loyal users" in business generally relies on the usage frequency of users over a period of time. Most of the early research on loyal users was based on usage behavior, defining repeated use more than 3 times consecutively as loyalty.

[0003] It should be noted that the information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, and thus may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0004] Embodiments of the present disclosure provide a method and apparatus for classifying users, a computer-readable storage medium, and an electronic device, which can achieve the classification of users.

[0005] Other features and advantages of the present disclosure will become apparent through the following detailed description, or be learned in part through the practice of the present disclosure.

[0006] According to one aspect of the present disclosure, there is provided a method for classifying users, including:

[0007] When obtaining user authorization, obtaining a plurality of user behavior information within a first time period, where each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior;

[0008] Obtaining, according to the plurality of user behavior information, the time, the number of behaviors, and the total resources used in the most recent behavior of the first category of each user;

[0009] Obtaining, according to the time, the number of behaviors, and the total resources used in the most recent behavior of the first category of each user, a time threshold, a behavior number threshold, and a total resources information threshold for the most recent behavior of the first category of the plurality of users, or obtaining, according to the time, the number of behaviors of the first category of each user, and the plurality of user behavior information, a time threshold, a behavior number threshold, and a total resources information threshold for the most recent behavior of the first category of the plurality of users;

[0010] Among the multiple users, users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior count threshold, and the information on the total resources used for the behavior is greater than or equal to the information threshold of the total resources used for the behavior are identified as first-level users of the first category.

[0011] In one embodiment, obtaining the time threshold, behavior count threshold, and information threshold of the total resources used for the most recent behavior of the first category of the multiple users based on the time, number of behaviors, and information on the total resources used for the most recent behavior of the first category of each user includes:

[0012] Obtaining the time threshold and behavior count threshold of the most recent behavior of the first category of the multiple users by clustering based on the time and number of behaviors of the most recent behavior of the first category of each user among the multiple users;

[0013] Determining the information threshold of the total resources used for the behavior of the first category of the multiple users based on the time threshold and behavior count threshold of the most recent behavior of the first category of the multiple users.

[0014] In one embodiment, obtaining the time threshold and behavior count threshold of the most recent behavior of the first category of the multiple users by clustering based on the time and number of behaviors of the most recent behavior of the first category of each user among the multiple users includes:

[0015] Dividing the multiple users into 4 regions by clustering based on the time and number of behaviors of the most recent behavior of the first category of each user among the multiple users;

[0016] Taking the minimum user behavior count in the region with the highest average number of behaviors among the 4 regions as the behavior count threshold;

[0017] Taking the median value of the time of the most recent behavior of the users in the region with the second-highest average number of behaviors among the 4 regions as the time threshold of the most recent behavior.

[0018] In one embodiment, determining the information threshold of the total resources used for the behavior of the first category of the multiple users based on the time threshold and behavior count threshold of the most recent behavior of the first category of the multiple users includes:

[0019] Dividing the first type of users and the second type of users based on the time threshold and behavior count threshold of the most recent behavior of the first category of the multiple users;

[0020] Determining the information threshold of the total resources used for the behavior of the multiple users based on the distribution diagrams of the first type of users and the second type of users.

[0021] In one embodiment, determining the information threshold of the total resources used by the behaviors of the multiple users according to the distribution maps of the first type of users and the second type of users includes:

[0022] Forming a first distribution map according to the information of the total resources used by the behaviors of the first type of users and the proportion of the information of the total resources used by the behaviors;

[0023] Forming a second distribution map according to the information of the total resources used by the behaviors of the second type of users and the proportion of the information of the total resources used by the behaviors;

[0024] Taking the intersection point of the boundary curves of the first distribution map and the second distribution map as the information threshold of the total resources used by the behaviors of the first category of the multiple users.

[0025] In one embodiment, obtaining the time threshold, the behavior frequency threshold, and the information threshold of the total resources used by the behaviors of the first category of the multiple users according to the time, the behavior frequency, and the multiple user behavior information of the most recent behavior of the first category of each user includes:

[0026] When the behavior frequency of the first category of each user among the multiple users is equal to 1, setting the behavior frequency threshold of the first category of the multiple users to 1, setting the time of the most recent behavior of the first category of the multiple users to L, and setting the time threshold of the most recent behavior of the first category of the multiple users to L, where L is a positive number greater than or equal to 0;

[0027] Obtaining the total number of other third-level categories under the same second-level category of the first category and the total information of the resources used by the behaviors of each category of the other third-level categories and the information threshold of the total resources used by the behaviors according to the multiple user behavior information;

[0028] Obtaining the information threshold of the total resources used by the behaviors of the first category of the multiple users according to the total number of other third-level categories under the same second-level category of the first category and the total information of the resources used by the behaviors of each category of the other third-level categories and the information threshold of the total resources used by the behaviors.

[0029] In one embodiment, the method further includes:

[0030] Removing abnormal users from the multiple users according to the user identification.

[0031] According to one aspect of the present disclosure, there is provided a division device for user classification, including:

[0032] The first acquisition module is configured to acquire a plurality of user behavior information within a first time period when obtaining user authorization, where each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior;

[0033] The second acquisition module is configured to obtain, according to the plurality of user behavior information, the time, the number of behaviors, and the information on the total resources used in the most recent behavior of the first category of each user;

[0034] The third acquisition module is configured to obtain, according to the time, the number of behaviors, and the information on the total resources used in the most recent behavior of the first category of each user, the time threshold, the behavior number threshold, and the information threshold of the total resources used in the most recent behavior of the first category of the plurality of users, or to obtain, according to the time, the number of behaviors of the most recent behavior of the first category of each user, and the plurality of user behavior information, the time threshold, the behavior number threshold, and the information threshold of the total resources used in the most recent behavior of the first category of the plurality of users;

[0035] The determination module is configured to confirm, among the plurality of users, the users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior number threshold, and the information on the total resources used in the behavior is greater than or equal to the information threshold of the total resources used in the behavior as the first-level users of the first category.

[0036] According to one aspect of the present disclosure, there is provided an electronic device, including:

[0037] One or more processors;

[0038] A storage device configured to store one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the method described in any one of the above embodiments.

[0039] According to one aspect of the present disclosure, there is provided a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the method described in any one of the above embodiments.

[0040] The present disclosure proposes a method that integrates information on the time of the most recent behavior, the number of behaviors, and the total resources used in the behavior (RF-M model) to divide users from a cross-perspective, breaking through the limitations of traditional methods that only consider a single indicator, and the division results are more meaningful for reference.

[0041] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present disclosure. Description of the Drawings

[0042] The following drawings depict certain illustrative embodiments of the present invention, where like reference numerals represent like elements. These described embodiments will be exemplary embodiments of the present disclosure and are not limiting in any way.

[0043] Figure 1 A schematic diagram of an exemplary system architecture showing a partitioning method for user classification to which embodiments of the present disclosure can be applied;

[0044] Figure 2 A schematic diagram of the structure of a computer system of an electronic device suitable for implementing embodiments of the present disclosure;

[0045] Figure 3 A flowchart schematically showing a partitioning method for user classification according to an embodiment of the present disclosure;

[0046] Figure 4 A schematic diagram of the operating system structure for partitioning user classification according to an embodiment of the present disclosure;

[0047] Figure 5 A schematic diagram of the process for a partitioning method of user classification according to an embodiment of the present disclosure;

[0048] Figure 6 It is a distribution diagram of R and F data according to an embodiment of the present disclosure;

[0049] Figure 7 It is a schematic distribution diagram after clustering of R and F data according to an embodiment of the present disclosure;

[0050] Figure 8 It is a schematic distribution diagram of threshold partitioning after clustering of R and F data according to an embodiment of the present disclosure;

[0051] Figure 9 It is a distribution diagram of high-quality users and non-high-quality users M according to an embodiment of the present disclosure;

[0052] Figure 10 It is a schematic diagram of the threshold of the number of behaviors according to an embodiment of the present disclosure;

[0053] Figure 11 A block diagram schematically showing a partitioning device for user classification according to an embodiment of the present disclosure;

[0054] Figure 12 A block diagram schematically showing a partitioning device for user classification according to another embodiment of the present disclosure;

[0055] Figure 13 A block diagram schematically showing a partitioning device for user classification according to another embodiment of the present disclosure. Detailed Implementation Manner

[0056] Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art.

[0057] In addition, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of the embodiments of the present disclosure. However, those skilled in the art will realize that the technical solutions of the present disclosure can be practiced without one or more of the specific details, or other methods, components, devices, steps, etc. may be employed. In other cases, well-known methods, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the present disclosure.

[0058] The block diagrams shown in the drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor devices and / or microcontroller devices.

[0059] The flowcharts shown in the drawings are merely illustrative and do not necessarily include all the content and operations / steps, nor are they necessarily executed in the order described. For example, some operations / steps can be decomposed, while some operations / steps can be combined or partially combined, so the actual execution order may change according to the actual situation.

[0060] Figure 1 A schematic diagram of an exemplary system architecture 100 for a method of dividing user classification to which the embodiments of the present disclosure can be applied is shown.

[0061] As Figure 1 shown, the system architecture 100 may include one or more of terminals 101, 102, 103, a network 104, and a server 105. The network 104 is a medium for providing a communication link between the terminals 101, 102, 103 and the server 105. The network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, etc.

[0062] It should be understood that Figure 1 the number of terminals, networks, and servers in

[0063] Staff members can use terminals 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Terminals 101, 102, and 103 can be various electronic devices with display screens, including but not limited to smartphones, tablets, portable computers, desktop computers, digital movie projectors, and so on.

[0064] Server 105 can be a server that provides various services. For example, when a staff member sends a division request for user classification to the server through terminal 103 (which can also be terminal 101 or 102), when the server 105 obtains user authorization, it acquires multiple user behavior information within a first time period. Each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior; based on the multiple user behavior information, it obtains the time, number of behaviors, and total resources used in the most recent behavior of the first category for each user; based on the time, number of behaviors, and total resources used in the most recent behavior of the first category for each user, it obtains the time threshold, behavior count threshold, and total resources used information threshold for the most recent behavior of the first category of the multiple users, or based on the time, number of behaviors of the most recent behavior of the first category for each user and the multiple user behavior information, it obtains the time threshold, behavior count threshold, and total resources used information threshold for the most recent behavior of the first category of the multiple users; it identifies users among the multiple users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior count threshold, and the information on the total resources used is greater than or equal to the total resources used information threshold as first-level users of the first category. Server 105 can display the division result of user classification on terminal 103 or other terminals, and then staff members or other personnel can view the division result of user classification based on the content displayed on the terminal.

[0065] For another example, the terminal 103 (which can also be the terminal 101 or 102) can be a smart TV, a VR (Virtual Reality) / AR (Augmented Reality) helmet display, or a mobile terminal such as a smart phone or a tablet computer with navigation, online car-hailing, instant messaging, video application (APP) and so on installed thereon. A staff member can submit a request for dividing user categories through the smart TV, the VR / AR helmet display, or the navigation, online car-hailing, instant messaging, video APP. The server 105 can obtain the division result of the user categories. The server 105 can return the obtained division result of the user categories to the smart TV, the VR / AR helmet display, or the navigation, online car-hailing, instant messaging, video APP, and then display the obtained division result of the user categories through the smart TV, the VR / AR helmet display, or the navigation, online car-hailing, instant messaging, video APP.

[0066] Figure 2 FIG. shows a schematic structural diagram of a computer system of an electronic device suitable for implementing the embodiments of the present disclosure.

[0067] It should be noted that Figure 2 The computer system 200 of the electronic device shown is only an example and should not impose any limitation on the functions and the scope of use of the embodiments of the present disclosure.

[0068] As Figure 2 shown, the computer system 200 includes a central processing unit (CPU, Central Processing Unit) 201, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM, Read-Only Memory) 202 or the program loaded from the storage section 208 into the random access memory (RAM, Random Access Memory) 203. In the RAM 203, various programs and data required for system operation are also stored. The CPU 201, the ROM 202, and the RAM 203 are connected to each other through a bus 204. The input / output (I / O) interface 205 is also connected to the bus 204.

[0069] The following components are connected to the I / O interface 205: an input section 206 including a keyboard, a mouse, etc.; an output section 207 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 208 including a hard disk, etc.; and a communication section 209 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 209 performs communication processing via a network such as the Internet. A drive 210 is also connected to the I / O interface 205 as needed. A removable medium 211, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 210 as needed so that a computer program read therefrom is installed into the storage section 208 as needed.

[0070] Specifically, according to an embodiment of the present disclosure, the processes described below with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present disclosure includes a computer program product that includes a computer program carried on a computer-readable storage medium, and the computer program includes program code for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network via the communication section 209, and / or installed from the removable medium 211. When the computer program is executed by a central processing unit (CPU) 201, various functions defined in the methods and / or apparatuses of the present application are executed.

[0071] It should be noted that the computer-readable storage medium shown in the present disclosure can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM (Erasable Programmable Read Only Memory)), or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, the computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program code is carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable storage medium other than the computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wireless, wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0072] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods, apparatuses, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions boxed in the block can occur in a different order than that boxed in the accompanying drawings. For example, two consecutive blocks shown can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0073] The modules and / or units and / or subunits involved in the embodiments of the present disclosure can be implemented in software or in hardware, and the described modules and / or units and / or subunits can also be provided in a processor. Among them, the names of these modules and / or units and / or subunits do not constitute a limitation to the modules and / or units and / or subunits themselves in some cases.

[0074] On the other hand, the present application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The above computer-readable storage medium carries one or more programs, and when the one or more programs are executed by an electronic device, the electronic device implements the method described in the following embodiments. For example, the electronic device can implement as Figure 3 each step.

[0075] In related technologies, for example, machine learning methods, deep learning methods, etc. can be used to classify users, and different methods are applicable to different ranges.

[0076] Figure 3 Schematically shows a flowchart of a method for classifying users according to an embodiment of the present disclosure. The method steps of the embodiments of the present disclosure can be executed by a terminal or a server, or by interaction between the terminal and the server. For example, it can be executed by the server 105 in the above Figure 1 , but the present disclosure is not limited thereto.

[0077] In step S310, when obtaining user authorization, obtain a plurality of user behavior information within a first time period, where each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior.

[0078] In this step, when the server obtains user authorization, it obtains a plurality of user behavior (such as consumption) information within a first time period, where each user behavior information includes a user identification code (PIN, Personal Identification Number), the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior (the information on the resources used, for example, the amount consumed).

[0079] In one embodiment, the first time period can be any period of time, such as 1 month, 1 year, or 2 years, etc.

[0080] In the embodiments of the present disclosure, the terminal can be implemented in various forms. For example, the terminal described in the present disclosure may include mobile terminals such as mobile phones, tablet computers, laptop computers, handheld computers, personal digital assistants (PDAs), portable media players (PMPs), user classification devices, wearable devices, smart bracelets, pedometers, robots, driverless vehicles, etc., and fixed terminals such as digital TVs (televisions) and desktop computers.

[0081] In step S320, based on the multiple user behavior information, obtain the time, number of behaviors, and information on the total resources used in the behavior (total amount consumed) of the most recent behavior of each user in the first category.

[0082] In this step, the server obtains the time, number of behaviors, and information on the total resources used in the behavior of the most recent behavior of each user in the first category based on the multiple user behavior information.

[0083] In one embodiment, the first category can be any product category, and the category can include the division of first-level categories, second-level categories, and third-level categories, such as computer office - electronic equipment - monitors, etc.

[0084] In step S330, obtain the time threshold, number of behavior threshold, and information threshold of the total resources used in the behavior of the most recent behavior of the multiple users in the first category based on the time, number of behaviors, and information on the total resources used in the behavior of the most recent behavior of each user in the first category, or obtain the time threshold, number of behavior threshold, and information threshold of the total resources used in the behavior of the most recent behavior of the multiple users in the first category based on the time, number of behaviors of the most recent behavior of each user in the first category, and the multiple user behavior information.

[0085] In this step, the server obtains the time threshold, number of behavior threshold, and information threshold of the total resources used in the behavior of the most recent behavior of the multiple users in the first category based on the time, number of behaviors, and information on the total resources used in the behavior of the most recent behavior of each user in the first category, or obtains the time threshold, number of behavior threshold, and information threshold of the total resources used in the behavior of the most recent behavior of the multiple users in the first category based on the time, number of behaviors of the most recent behavior of each user in the first category, and the multiple user behavior information.

[0086] In step S340, users among the multiple users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior number threshold, and the information on the total resources used by the behavior is greater than or equal to the information threshold of the total resources used by the behavior are identified as first-level users of the first category.

[0087] In this step, the server identifies, among the multiple users, users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior number threshold, and the information on the total resources used by the behavior is greater than or equal to the information threshold of the total resources used by the behavior as first-level users of the first category (first-level users are, for example, loyal users).

[0088] Among them, the time of the most recent behavior being less than or equal to the time threshold of the most recent behavior means that: compared with the time threshold of the most recent behavior, the time of the most recent behavior is closer to or the same as the classification time.

[0089] In the present disclosure, by integrating the information of the most recent usage time, usage frequency, and total resources used by the behavior (RF-M model), users are divided from a cross-perspective, breaking through the limitations of traditional methods that only consider a single indicator, and the division results are more meaningful for reference.

[0090] In the RFM model, R (Recency) refers to the time of the most recent use, F (Frequency) refers to the number of uses within a period of time, and M (Monetary) refers to the information on the resources used by the behavior within a period of time.

[0091] (1) The following introduces the specific implementation of the clustering method in the classification and division method:

[0092] In one embodiment, obtaining the time threshold of the most recent behavior, the behavior number threshold, and the information threshold of the total resources used by the behavior of the first category of the multiple users according to the time of the most recent behavior, the number of behaviors, and the information on the total resources used by the behavior of the first category of each user includes:

[0093] Obtaining the time threshold of the most recent behavior and the behavior number threshold of the first category of the multiple users by means of clustering according to the time of the most recent behavior and the number of behaviors of each user in the multiple users;

[0094] Determining the information threshold of the total resources used by the behavior of the first category of the multiple users according to the time threshold of the most recent behavior and the behavior number threshold of the first category of the multiple users.

[0095] In one embodiment, obtaining the time threshold and the behavior frequency threshold of the most recent behavior of the first category for each user among the multiple users according to the time and the number of behaviors of the most recent behavior of the first category for each user among the multiple users through a clustering method includes:

[0096] Dividing the multiple users into 4 regions through a clustering method according to the time and the number of behaviors of the most recent behavior of the first category for each user among the multiple users;

[0097] Taking the minimum user behavior frequency in the region with the highest average behavior frequency among the 4 regions as the behavior frequency threshold;

[0098] Taking the median value of the time of the most recent behavior of the users in the region with the second highest average behavior frequency among the 4 regions as the time threshold of the most recent behavior.

[0099] In one embodiment, determining the information threshold of the total resources used by the behavior of the first category for the multiple users according to the time threshold and the behavior frequency threshold of the most recent behavior of the first category for the multiple users includes:

[0100] Dividing the first type of users (for example, high-quality users) and the second type of users (for example, non-high-quality users) according to the time threshold and the behavior frequency threshold of the most recent behavior of the first category for the multiple users;

[0101] Determining the information threshold of the total resources used by the behavior of the multiple users according to the distribution diagrams of the first type of users and the second type of users.

[0102] In one embodiment, determining the information threshold of the total resources used by the behavior of the multiple users according to the distribution diagrams of the first type of users and the second type of users includes:

[0103] Forming a first distribution diagram according to the information of the total resources used by the behavior of the first type of users and the proportion of the information of the total resources used by the behavior;

[0104] Forming a second distribution diagram according to the information of the total resources used by the behavior of the second type of users and the proportion of the information of the total resources used by the behavior;

[0105] Taking the intersection point of the boundary curves of the first distribution diagram and the second distribution diagram as the information threshold of the total resources used by the behavior of the first category for the multiple users.

[0106] (II) The following introduces the specific implementation manners of the classification method for special category products in the classification method:

[0107] In one embodiment, obtaining the time threshold, the behavior frequency threshold, and the information threshold of the total resources used for the most recent behavior of the first category of the multiple users based on the time, the number of behaviors, and the multiple user behavior information of the first category of each user includes:

[0108] When the number of behaviors of the first category of each user among the multiple users is equal to 1, set the behavior frequency threshold of the first category of the multiple users to 1, set the time of the most recent behavior of the first category of the multiple users to L, and set the time threshold of the most recent behavior of the first category of the multiple users to L, where L is a positive number greater than or equal to 0;

[0109] Obtain the total number of other third-level categories under the same second-level category of the first category, the total information of the resources used for the behavior of each category of the other third-level categories, and the information threshold of the total resources used for the behavior based on the multiple user behavior information;

[0110] Obtain the information threshold of the total resources used for the behavior of the first category of the multiple users based on the total number of other third-level categories under the same second-level category of the first category, the total information of the resources used for the behavior of each category of the other third-level categories, and the information threshold of the total resources used for the behavior.

[0111] In one embodiment, obtaining the information threshold of the total resources used for the behavior of the first category of the multiple users based on the total number of other third-level categories under the same second-level category of the first category, the total information of the resources used for the behavior of each category of the other third-level categories, and the information threshold of the total resources used for the behavior includes:

[0112] Obtain the information threshold of the total resources used for the behavior of the first category according to the following formula:

[0113]

[0114] where M is the M threshold (information threshold of the total resources used for the behavior) of the first category to be calculated;

[0115] N is the total number of other third-level categories under the same second-level category as the first category;

[0116] GMV i is the total score value (total information of the resources used for the behavior) of category i;

[0117] GMV j is the total score value of category j;

[0118] M i is the M threshold (information threshold of the total resources used for the behavior) of category i in other third-level categories.

[0119] In one embodiment, the method further includes: removing abnormal users from the multiple users according to user identification. Abnormal users include, for example, brush order users, enterprise purchase users, distribution users, etc.

[0120] (3) The classification method of the present application will be described in detail in combination with specific steps as follows:

[0121] In the prior art, repeated use more than 3 times in a row is defined as first-level users (loyal users). However, in actual operation, there are a wide variety of product categories, and the reuse (repeated consumption) cycles are different. Such a "one-size-fits-all" method cannot adapt to different categories. For example: the reuse frequency of users in some product categories is relatively high, while the reuse frequency of users in slow-selling products such as raw diamonds and jade is often very low. Moreover, this method only considers the usage frequency of users, while the actual discrimination of user classification is often reflected in aspects such as the information of resources used in the behavior and the recent usage time. The present disclosure introduces an algorithm that fuses multiple models (from two aspects of RFM and clustering) for user classification under multiple categories. An RFM model is established to classify users from the cross-level of multiple perspectives. And in combination with the specific user marketing needs of the large platform, in thousands of categories of the entire platform, in an unsupervised learning (such as clustering) manner, loyal users are adaptively identified under each category.

[0122] The present disclosure introduces the RF-M model to divide users from a cross-perspective, overcoming the limitation of the existing first-level (loyal) user identification method that only considers a single indicator of repeat purchase. Compared with the traditional user division strategy, the present disclosure innovatively integrates unsupervised learning, enabling the model to adaptively identify first-level users under different categories. Avoiding the accuracy loss caused by the "one-size-fits-all" scheme in different categories.

[0123] The traditional strategy can only produce division results scene by scene. The present invention optimizes the system performance, can obtain the user division thresholds under thousands of categories at one time, achieving the effect of batch processing of a large number of user groups in multiple categories and greatly improving the user division speed.

[0124] Figure 4 Shows a schematic diagram of the division operation system structure of user classification in an embodiment of the present disclosure. Refer to Figure 4 , including:

[0125] Data integration: Extract order data and product data of the e-commerce platform from the big data mart, and integrate the captured data into Hadoop (distributed system infrastructure) for storage. Clean up interference data, and eliminate data such as enterprise purchase orders that affect the model results.

[0126] Data calculation: Perform adaptive calculations. Through unsupervised learning, learn the labels of each user under different perspectives for each category. The information of the user PIN (Personal Identification Number) will be matched with all the categories that the user has used (consumed). At the same time, the order details are also associated with the category-user combination. Therefore, distributed calculations can be performed on the data under different categories.

[0127] Data processing: Perform data distribution verification to identify special categories where the tagging results are abnormal due to the data distribution being concentrated on the same observed value. Transfer the categories with the same F value for all users to the special category processing step and re-tag the users under the special categories.

[0128] Business delivery: Store the user classification results under thousands of categories calculated in batches by the model in Hadoop. In actual business, by judging the relationship between the actual usage behavior of the user and the calculation threshold of the category RFM model, it can be determined which sub-operation type the user belongs to, so as to obtain the user group PIN package under the sub-operation scenario.

[0129] The system design of this disclosure combines the RFM model with unsupervised learning to achieve model improvement that can be adaptively processed from multiple perspectives in batches. In the RFM model, R (Recency) refers to the time of the last use (consumption), F (Frequency) refers to the number of uses (consumptions) within a period of time, and M (Monetary) refers to the information of the total resources used (total consumption amount) for the behavior within a period of time. These three perspectives respectively describe the timeliness, activity, and value contribution of users. As Figure 4 shown, this operating system designs 5 modules: data integration, data calculation, data processing, business delivery, and system management. The solutions are described in detail as follows:

[0130] In the data integration step, the partitioning system obtains data such as orders, users, and categories (categories such as food, beverages, and household appliances, etc.) from the big data mart and integrates them for storage on Hadoop. Clean the data. Since in addition to ordinary category users, there are also some special users, such as brush order users, enterprise purchase users, distribution users, etc. Taking brush order users as an example, the characteristics of such users are high timeliness, high activity, and high value contribution, but they are not a real first-level user. Therefore, this disclosure needs to remove such users and retain the most authentic ordinary data for identification. Special users have been marked in other systems. This application does not consider how to distinguish and obtain them, but directly identifies and uses them according to the marks.

[0131] In the next step, the assembled data will be transferred to the data calculation and data processing steps of the partitioning system. Based on the order wide table, information such as user PIN, usage time, and resources used in behavior within a certain period (for example, two years) is retrieved to calculate the user R, F, and M values, obtaining data in the dimensions of category, user, R, F, and M. In the RF stage, according to the user R and F data, clustering is performed to distinguish high-quality users (the first type of users) and non-high-quality users (the second type of users), and the R and F thresholds are determined. In the M stage, the M data of high-quality users and non-high-quality users are made into a histogram, and the M threshold is determined according to its approximate Gaussian distribution. In the next step, the system will transfer the calculated data to the special category identification step: although the R, F, and M thresholds have been determined in the previous steps, for some categories, the usage frequency of all users is 1 time. In this case, the discrimination of loyal users has no differentiability in the dimension of usage frequency, so the usage frequency will no longer participate in the discrimination of loyal users, and such categories must be specially processed. In the last step of identification and marking, according to the R, F, and M of the category, it is judged whether each user is a loyal user (level 1 user). The specific process is as Figure 5 。 Figure 5 shows a flowchart of the partitioning method for user classification according to an embodiment of the present disclosure.

[0132] The structure of the integrated data table is as follows. The time span is, for example, two years, such as from January 2019 to December 2020:

[0133]

[0134]

[0135] RF stage: According to business understanding, users have a life cycle. Level 1 users should belong to high-value users, that is, they have used recently and have a high usage frequency. Figure 6 is the R, F data distribution diagram according to an embodiment of the present disclosure. The R, F data distribution is as Figure 6 , bottom left: used recently, low usage frequency, potential users. Top left: used recently, high usage frequency, high-value users. Bottom right: not used recently, low usage frequency, pre-lost users. Figure 6 In, the unit of R is times, and the unit of F is days.

[0136] Figure 7 is the schematic distribution diagram after clustering of the R, F data according to an embodiment of the present disclosure.

[0137] According to the business, the roles of users are fluid. Based on the R, F data, users are divided into four regions, and the effect is as Figure 7 , with the abscissa being R and the ordinate being F. The unit of R is times, and the unit of F is days. The "first region" part is high-value users.

[0138] Figure 8It is a schematic distribution diagram of threshold division after R and F data clustering in an embodiment of the present disclosure.

[0139] Figure 8 Users have been divided into high-quality users (the first region) and non-high-quality users (the region other than the first region) to determine the R and F thresholds. As Figure 8 shown, the lower boundary line Q of the first region is the threshold of F; determine the midpoint of the third region, and the straight line P perpendicular to the X-axis passing through the midpoint of the third region is the threshold of R.

[0140] Stage M: In the second step, the R and F thresholds of the category have been obtained. If r <= R and f >= F, they are high-quality users, and the rest are non-high-quality users. Figure 9 It is a distribution diagram of M for high-quality users and non-high-quality users in an embodiment of the present disclosure. Obtain the M data of the two. It is found through observation that there are obvious differences in the M distributions of high-quality users and non-high-quality users (taking Figure 9 as an example, the M value of non-high-quality users ranges from 0 to 400, and the M value of high-quality users ranges from 200 to 800), and the effect is as Figure 9 shown. Figure 10 It is a schematic diagram of the behavior frequency threshold in an embodiment of the present disclosure. Make a bar chart of M for high-quality users and non-high-quality users, and it is found that the graph generally follows a skewed distribution. As Figure 10 shown, non-high-quality users (the second distribution diagram), high-quality users (the first distribution diagram). Take the intersection point of the boundary curves of the first distribution diagram and the second distribution diagram as the information threshold of the total resources used by the behavior of the first category of the multiple users, that is, the intersection point A is the threshold of M. Figure 10 The unit of the abscissa is yuan, and the unit of the ordinate is the proportion of the information of the resources used by the behavior.

[0141] Special category processing steps: There are some special categories. For example, if the usage frequency of all users is 1 time, then there is no discrimination in the dimension of usage frequency for loyal user discrimination, and the usage frequency no longer participates in the discrimination of loyal users. Such categories must be specially processed. The processing method is to distinguish on M and use other third-level categories under the same second-level category for filling. The reason for this processing is that the categories under the same second-level category are relatively similar. The specific formula is as follows:

[0142]

[0143] M: The M threshold of the special category to be calculated (the information threshold of the total resources used by the behavior);

[0144] N: The total number of other third-level categories under the same second-level category as the special category;

[0145] GMV i : The total score of category i (the total information of the resources used by the behavior);

[0146] GMV j : The total score of category j;

[0147] M i : The M threshold of category i in other third-level categories (the information threshold of the total resources used by the behavior).

[0148] Identification and marking steps: After the RF stage, M stage, and special category processing steps in the system, the R, F, and M thresholds of each level of category have been determined. If the user meets the conditions of r <= R, f >= F, and m >= M, it is determined as a loyal user (level 1 user), otherwise it is a non-loyal user. Different from the previous solutions that need to output the division results one by one for each analysis scenario, this system can batch calculate the full-scale user marking results under thousands of sub-category scenarios and push them to the application side for the business to extract the user PIN package at any time.

[0149] Feedback loop: Users can master the distribution form and user group ratio of different user groups under this category through the visual user segmentation image. For the situation where there is a certain deviation between the local user marking effect and the actual business, or the user group ratio is unbalanced, the threshold can be optimized and adjusted. For example, smooth the group boundary points or directly adjust the threshold reference point. Thus, the business cognition is input back to the system model to form a dynamic feedback loop.

[0150] From the perspective of user division, this disclosure combines the RF-M model and clustering algorithm to divide users from a cross-perspective, breaking through the limitation of traditional methods that only consider a single indicator. The usage frequencies of users in different categories and the information of the resources used by their behaviors are different. Utilizing the characteristics of unsupervised learning, the division threshold is adaptively obtained under categories with different characteristics. A feedback mechanism is provided, and each division has a systematic optimization and improvement for the next time. This invention integrates data reading, calculation, storage, and application functions, systematically batch processes the user division solutions under thousands of categories, and directly outputs the results of category-user PIN-user marking to the application side. The traditional division scheme process is to manually formulate division rules according to user activity and contribution degree in each scenario. The improved scheme is more suitable for scenarios such as large-scale user hierarchical marketing in e-commerce, greatly improving the system production efficiency.

[0151] Figure 11 The block diagram of the division device for user classification according to an embodiment of the present disclosure is schematically shown. The division device 1100 for user classification provided by the embodiment of the present disclosure can be set on the server side or the terminal, or part of it can be set on the terminal and part on the server side. For example, it can be set in Figure 1 the server 105 in, but the present disclosure is not limited thereto.

[0152] The classification device 1100 for user classification provided by the embodiments of the present disclosure may include a first acquisition module 1110, a second acquisition module 1120, a third acquisition module 1130, and a determination module 1140.

[0153] Among them, the first acquisition module is configured to, when obtaining user authorization, acquire a plurality of user behavior information within a first time period, where each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior;

[0154] The second acquisition module is configured to acquire, according to the plurality of user behavior information, the time, the number of behaviors, and the information on the total resources used in the most recent behavior of each user for a first category;

[0155] The third acquisition module is configured to acquire, according to the time, the number of behaviors, and the information on the total resources used in the most recent behavior of each user for a first category, a time threshold, a number-of-behaviors threshold, and an information threshold on the total resources used in the most recent behavior of the plurality of users, or to acquire, according to the time, the number of behaviors, and the plurality of user behavior information of each user for a first category, a time threshold, a number-of-behaviors threshold, and an information threshold on the total resources used in the most recent behavior of the plurality of users;

[0156] The determination module is configured to identify, among the plurality of users, users whose time of the most recent behavior for the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the number-of-behaviors threshold, and the information on the total resources used in the behavior is greater than or equal to the information threshold on the total resources used in the behavior as first-level users of the first category.

[0157] According to the device embodiment of the present disclosure, by integrating information on the time of the most recent behavior, the number of behaviors, and the total resources used in the behavior (RF-M model), users are classified from a cross-perspective, breaking through the limitations of traditional methods that only consider a single indicator.

[0158] Figure 12 A block diagram of a classification device 1200 for user classification according to another embodiment of the present disclosure is schematically shown.

[0159] As Figure 12 shown, in addition to Figure 11 the first acquisition module 1110, the second acquisition module 1120, the third acquisition module 1130, and the determination module 1140 described in the embodiment, the classification device 1200 for user classification further includes a display module 1210.

[0160] Specifically, after the determination module 1140 determines the user classification, the display module 1210 displays the classification result of the user classification to the staff.

[0161] In the user classification division device 1200, the division result of the user classification can be displayed through the display module 1210.

[0162] Figure 13 A block diagram of a user classification division device 1300 according to another embodiment of the present disclosure is schematically shown.

[0163] As Figure 13 shown, in addition to Figure 11 the first acquisition module 1110, the second acquisition module 1120, the third acquisition module 1130, and the determination module 1140 described in the embodiment, the user classification division device 1300 further includes a storage module 1310.

[0164] Specifically, the storage module 1310 is used to store data during the division process of the user classification for convenient subsequent invocation and reference.

[0165] It can be understood that the first acquisition module 1110, the second acquisition module 1120, the third acquisition module 1130, the determination module 1140, the display module 1210, and the storage module 1310 can be implemented in one module, or any one of the modules can be split into multiple modules. Or, at least part of the functions of one or more of these modules can be combined with at least part of the functions of other modules and implemented in one module. According to an embodiment of the present disclosure, at least one of the first acquisition module 1110, the second acquisition module 1120, the third acquisition module 1130, the determination module 1140, the display module 1210, and the storage module 1310 can be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on chip, a system on substrate, a system on package, an application specific integrated circuit (ASIC), or can be implemented in any other reasonable way of integrating or packaging circuits, etc., in hardware or firmware, or implemented in an appropriate combination of software, hardware, and firmware. Or, at least one of the first acquisition module 1110, the second acquisition module 1120, the third acquisition module 1130, the determination module 1140, the display module 1210, and the storage module 1310 can be at least partially implemented as a computer program module, and when the program is run by a computer, the functions of the corresponding module can be executed.

[0166] It should be noted that although several modules, units, and subunits of the device for action execution are mentioned in the above detailed description, this division is not mandatory. In fact, according to the embodiments of the present disclosure, the features and functions of two or more of the above-described modules, units, and subunits can be embodied in one module, unit, and subunit. Conversely, the features and functions of one module, unit, and subunit described above can be further divided and embodied by multiple modules, units, and subunits.

[0167] Through the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, and includes several instructions to enable a computing device (such as a personal computer, a server, a touch terminal, or a network device, etc.) to execute the method according to the embodiments of the present disclosure.

[0168] After considering the specification and practicing the disclosure herein, those skilled in the art will readily conceive of other embodiments of the present disclosure. This application is intended to cover any variations, uses, or adaptations of the present disclosure, which follow the general principles of the present disclosure and include known common knowledge or conventional technical means in the technical field not disclosed in the present disclosure. The specification and embodiments are only regarded as exemplary, and the true scope and spirit of the present disclosure are pointed out by the following claims.

[0169] It should be understood that the present disclosure is not limited to the exact structures described above and shown in the drawings, and various modifications and changes can be made without departing from its scope. The scope of the present disclosure is only limited by the appended claims.

Claims

1. A method for classifying users, characterized in that, Including: When obtaining user authorization, obtain multiple user behavior information within a first time period. Each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and information on the resources used in each behavior; According to the multiple user behavior information, obtain the time, the number of behaviors, and the information on the total resources used in the most recent behavior of the first category for each user; Obtain the time threshold, the behavior number threshold, and the information threshold on the total resources used in the most recent behavior of the first category of the multiple users according to the time, the number of behaviors, and the information on the total resources used in the most recent behavior of the first category for each user, or obtain the time threshold, the behavior number threshold, and the information threshold on the total resources used in the most recent behavior of the first category of the multiple users according to the time, the number of behaviors, and the multiple user behavior information of the most recent behavior of the first category for each user; Identify users among the multiple users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the number of behaviors is greater than or equal to the behavior number threshold, and the information on the total resources used in the behavior is greater than or equal to the information threshold on the total resources used in the behavior as first-level users of the first category; Among them, obtaining the time threshold, the behavior number threshold, and the information threshold on the total resources used in the most recent behavior of the first category of the multiple users according to the time, the number of behaviors, and the information on the total resources used in the most recent behavior of the first category for each user includes: Obtain the time threshold and the behavior number threshold of the most recent behavior of the first category of the multiple users by clustering according to the time and the number of behaviors of the most recent behavior of the first category for each user among the multiple users; Determine the information threshold on the total resources used in the behavior of the first category of the multiple users according to the time threshold and the behavior number threshold of the most recent behavior of the first category of the multiple users; Determining the information threshold on the total resources used in the behavior of the first category of the multiple users according to the time threshold and the behavior number threshold of the most recent behavior of the first category of the multiple users includes: Divide the multiple users into a first type of users and a second type of users according to the time threshold and the behavior number threshold of the most recent behavior of the first category of the multiple users; Determine the information threshold on the total resources used in the behavior of the multiple users according to the distribution diagrams of the first type of users and the second type of users.

2. The method according to claim 1, wherein Obtaining the time threshold and the behavior number threshold of the most recent behavior of the first category of the multiple users by clustering according to the time and the number of behaviors of the most recent behavior of the first category for each user among the multiple users includes: Divide the multiple users into 4 regions by clustering according to the time and the number of behaviors of the most recent behavior of the first category for each user among the multiple users; Use the minimum number of user behaviors in the region with the highest average number of behaviors among the 4 regions as the behavior number threshold; Use the median time of the most recent behavior of the users in the region with the second highest average number of behaviors among the 4 regions as the time threshold of the most recent behavior.

3. The method according to claim 2, characterized in that, Determining the information threshold of the total resources used by the behaviors of the multiple users according to the distribution maps of the first type of users and the second type of users includes: Forming a first distribution map according to the information of the total resources used by the behaviors of the first type of users and the proportion of the information of the total resources used by the behaviors; Forming a second distribution map according to the information of the total resources used by the behaviors of the second type of users and the proportion of the information of the total resources used by the behaviors; Taking the intersection point of the boundary curves of the first distribution map and the second distribution map as the information threshold of the total resources used by the behaviors of the first category of the multiple users.

4. The method according to claim 1, wherein Determining the time threshold, the behavior frequency threshold, and the information threshold of the total resources used by the behaviors of the first category of the multiple users according to the time, the behavior frequency, and the information of the total resources used by the behaviors of the first category of the most recent behaviors of each user and the multiple user behavior information includes: When the behavior frequency of the first category of each user among the multiple users is equal to 1, setting the behavior frequency threshold of the first category of the multiple users to 1 as well, setting the time of the most recent behavior of the first category of the multiple users to L, and setting the time threshold of the most recent behavior of the first category of the multiple users to L as well, where L is a positive number greater than or equal to 0; Obtaining the total number of other third-level categories under the same second-level category of the first category and the total information of the resources used by the behaviors of each category of the other third-level categories and the information threshold of the total resources used by the behaviors according to the multiple user behavior information; Obtaining the information threshold of the total resources used by the behaviors of the first category of the multiple users according to the total number of other third-level categories under the same second-level category of the first category and the total information of the resources used by the behaviors of each category of the other third-level categories and the information threshold of the total resources used by the behaviors.

5. The method according to claim 1, wherein Further includes: Removing abnormal users among the multiple users according to the user identification.

6. A classification device for user classification, characterized in that, Includes: A first acquisition module, configured to acquire multiple user behavior information within a first time period when user authorization is obtained, where each user behavior information includes a user identification code, the product category involved in each behavior, the time of each behavior, and the information of the resources used by each behavior; A second acquisition module, configured to acquire the time, the behavior frequency, and the information of the total resources used by the behaviors of the first category of each user according to the multiple user behavior information; A third acquisition module, configured to obtain the time threshold, the behavior count threshold, and the information threshold of the total resources used by the behavior of the most recent behavior of the first category of the multiple users according to the time, the behavior count, and the information of the total resources used by the behavior of the most recent behavior of the first category of each user, or to obtain the time threshold, the behavior count threshold, and the information threshold of the total resources used by the behavior of the most recent behavior of the first category of the multiple users according to the time, the behavior count, and the behavior information of the multiple users of the first category of each user; wherein, obtaining the time threshold, the behavior count threshold, and the information threshold of the total resources used by the behavior of the most recent behavior of the first category of the multiple users according to the time, the behavior count, and the information of the total resources used by the behavior of the most recent behavior of the first category of each user includes: obtaining the time threshold and the behavior count threshold of the most recent behavior of the first category of the multiple users by means of clustering according to the time and the behavior count of the most recent behavior of the first category of each user among the multiple users; determining the information threshold of the total resources used by the behavior of the first category of the multiple users according to the time threshold and the behavior count threshold of the most recent behavior of the first category of the multiple users; determining the information threshold of the total resources used by the behavior of the first category of the multiple users according to the time threshold and the behavior count threshold of the most recent behavior of the first category of the multiple users includes: dividing the multiple users into a first type of users and a second type of users according to the time threshold and the behavior count threshold of the most recent behavior of the first category of the multiple users; determining the information threshold of the total resources used by the behavior of the multiple users according to the distribution diagrams of the first type of users and the second type of users; A determination module, configured to confirm, among the multiple users, the users whose time of the most recent behavior of the first category is less than or equal to the time threshold of the most recent behavior, the behavior count is greater than or equal to the behavior count threshold, and the information of the total resources used by the behavior is greater than or equal to the information threshold of the total resources used by the behavior as the first-level users of the first category.

7. An electronic device, characterized in that, Comprising: One or more processors; A storage device, configured to store one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium storing a computer program, characterized in that, The computer program, when executed by a processor, implements the method according to any one of claims 1 to 5.

Images