User classification method, device and electronic equipment

Abstract

The present application discloses a user classification method, device and electronic device, which belongs to the field of communication technology and can solve the problem of low accuracy of user classification in the user operation mode in the prior art. The method includes: constructing at least one quality feature sequence and at least one original data feature sequence of the target user group, the target user group includes at least one user, each user corresponds to a quality feature sequence and an original data feature sequence, the quality feature sequence is used to indicate the quality of at least one business data, and the original data feature sequence is used to indicate the user's use of at least one business data; classifying the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavioral attribute feature group.

Description

Technical Field

[0001] The present application belongs to the field of communication technology, and specifically relates to a user classification method, device and electronic equipment. Background Art

[0002] With the continuous development of electronic devices, users have an increasing demand for intelligent recommendations on electronic devices (such as web page recommendations, shopping product recommendations, etc.). Therefore, developers need to upgrade the user operation method so that the content recommended by the electronic device meets the needs of users. However, the existing user classification method is based on user usage functions, or simply classifies users using the RFM basic model. In this way, the user operation method in the prior art has low accuracy in user classification. Summary of the invention

[0003] The purpose of the embodiments of the present application is to provide a user classification method, device and electronic device, which can solve the problem of low accuracy in user classification in the user operation mode in the prior art.

[0004] In order to solve the above technical problems, this application is implemented as follows:

[0005] In a first aspect, an embodiment of the present application provides a user classification method, which includes: constructing at least one quality feature sequence and at least one original data feature sequence for a target user group, the target user group includes at least one user, each user corresponds to a quality feature sequence and an original data feature sequence, the quality feature sequence is used to indicate the quality of at least one business data, and the original data feature sequence is used to indicate the user's usage of at least one business data; classifying the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavioral attribute feature group.

[0006] In a second aspect, an embodiment of the present application provides a user classification device, which includes: a construction module and a processing module. The construction module is used to construct at least one quality feature sequence and at least one original data feature sequence of a target user group, the target user group includes at least one user, each user corresponds to a quality feature sequence and an original data feature sequence, the quality feature sequence is used to indicate the quality of at least one business data, and the original data feature sequence is used to indicate the user's use of at least one business data; the processing module is used to classify the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavioral attribute feature group.

[0007] In a third aspect, an embodiment of the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the program or instructions are executed by the processor, the steps of the method described in the first aspect are implemented.

[0008] In a fourth aspect, an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented.

[0009] In a fifth aspect, an embodiment of the present application provides a chip, comprising a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement the method described in the first aspect.

[0010] In a sixth aspect, an embodiment of the present application provides a computer program product, which is stored in a storage medium and is executed by at least one processor to implement the method described in the first aspect.

[0011] In an embodiment of the present application, the electronic device can classify the target user group according to at least one quality feature sequence of the target user group, at least one original data feature sequence, and at least one behavioral attribute feature group of the user. In this solution, since the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for each user of the target user group, when the electronic device classifies the target user group, it can classify the target user group in combination with the quality of at least one piece of business data, the user's use of at least one piece of business data, and the user's behavioral attribute feature group, that is, the electronic device classifies the user group based on the quality of the business data, the use of the business data, and the user's behavioral attributes, so that the division of the user group is more accurate, thereby improving the accuracy of the electronic device in classifying users. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 This is one of the schematic diagrams of a user classification method provided in an embodiment of the present application;

[0013] Figure 2 This is a schematic diagram of an example of a user's behavior attribute feature group provided in an embodiment of the present application;

[0014] Figure 3 This is a second schematic diagram of a user classification method provided in an embodiment of the present application;

[0015] Figure 4 This is a schematic diagram of an example of attribute information of user classification provided by an embodiment of the present application;

[0016] Figure 5 This is a third schematic diagram of a user classification method provided in an embodiment of the present application;

[0017] Figure 6 is a structural diagram of a user classification device provided in an embodiment of the present application;

[0018] Figure 7 This is one of the hardware structure diagrams of an electronic device provided in an embodiment of the present application;

[0019] Figure 8 This is the second schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0021] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by "first", "second", etc. are generally of one type, and the number of objects is not limited. For example, the first object can be one or more. In addition, "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally indicates that the objects associated with each other are in an "or" relationship.

[0022] The user classification method provided in the embodiment of the present application is described in detail below through specific embodiments and application scenarios in conjunction with the accompanying drawings.

[0023] The user classification method in the embodiment of the present application can be applied to the scenario of dividing users in user operations.

[0024] At present, electronic devices can classify users according to product functions. User classification refers to dividing all users into groups with certain common characteristics according to different dimensions, and analyzing different user groups to find commonalities and differences. There are usually two ways to classify users. One solution is to simply classify according to user usage behavior, that is, those who use function A are classified into group A, and those who use function B are classified into group B. After classifying the main functions, the number of classified users determines which functions are the most popular among users, and the function value is determined according to the user value of the user group corresponding to the function point; another common solution is to cluster and classify according to simple user characteristics using a machine learning model. However, the existing user classification methods have shortcomings. For the first solution, when simply classifying according to user usage functions, the classification method is simple and the user value attribute information is insufficient; for the second solution, simple features are used to cluster users, which is not very explanatory. There is also a simple classification of users using the customer relationship management RFM basic model, but the rules are simple and the user attribute information is not fully utilized. In this way, the problem of inaccurate population division is faced in Internet business operations and effect optimization.

[0025] In the scheme provided in the embodiment of the present application, the electronic device can classify the target user group according to at least one quality feature sequence of the target user group, at least one original data feature sequence and at least one behavioral attribute feature group of the user. In this scheme, since the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for each user of the target user group, when the electronic device classifies the target user group, it can classify the target user group in combination with the quality of at least one business data, the user's use of at least one business data and the user's behavioral attribute feature group, that is, the electronic device classifies the user group based on the quality of the business data, the use of the business data and the user's behavioral attributes, so that the division of the user group is more accurate, thereby improving the accuracy of the electronic device in classifying users.

[0026] The present application embodiment provides a user classification method. Figure 1 FIG. 1 is a flowchart of a user classification method provided by an embodiment of the present application, which can be applied to electronic devices. Figure 1 As shown, the user classification method provided in the embodiment of the present application may include the following steps 201 and 202.

[0027] Step 201: The electronic device constructs at least one quality feature sequence and at least one original data feature sequence for a target user group.

[0028] In an embodiment of the present application, the above-mentioned target user group includes at least one user, each user corresponds to a quality feature sequence and an original data feature sequence, the quality feature sequence is used to indicate the quality of at least one business data, and the original data feature sequence is used to indicate the user's usage of at least one business data.

[0029] In an embodiment of the present application, the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for all users in the target user group, so that the electronic device can classify the target user group according to the quality feature sequence corresponding to each user in the target user group, the original data feature sequence and the user's behavioral attribute feature group to obtain at least one user group.

[0030] Optionally, in an embodiment of the present application, the target user group may be any one of the following: a user group of a shopping website (for example, all users who have registered on the shopping website), a user group of a reading website, a user group of a game application, a user group of a video application, etc.

[0031] Optionally, in an embodiment of the present application, "the electronic device constructs at least one quality feature sequence of the target user group" in the above step 201 can be specifically implemented through the following steps 201a and 201b.

[0032] Step 201a: The electronic device marks and scores at least one piece of business data to obtain at least one first label information.

[0033] In the embodiment of the present application, each first tag information includes an identifier of a piece of business data, a quality level to which a piece of business data belongs, and a quality score of a piece of business data.

[0034] In an embodiment of the present application, the electronic device can mark and score at least one piece of business data by training a classification model to obtain at least one first label information, and then the electronic device can determine a quality weight set based on the at least one first label information and a usage information, thereby obtaining a quality feature sequence.

[0035] Optionally, in an embodiment of the present application, the above-mentioned training classification model can be any one of the following: a decision tree model, w&d (fusion of a shallow (wide) model and a deep (deep) model).

[0036] Optionally, in the embodiment of the present application, at least one piece of business data may be marked and scored through manual annotation.

[0037] Optionally, in an embodiment of the present application, the at least one first tag information can be expressed as {w:a:b}, wherein w is an identifier of a piece of business data, a is a quality level to which a piece of business data belongs, and b is a quality score of a piece of business data.

[0038] Optionally, in an embodiment of the present application, the electronic device can pre-set multiple quality levels, for example: first level: low-level business data, second level: primary business data, third level: intermediate business data, fourth level: high-level business data, fifth level: superior business data, for subsequent determination of the quality level to which each piece of business data in at least one piece of business data belongs among the multiple quality levels.

[0039] Optionally, in an embodiment of the present application, the above-mentioned quality levels can be expressed by numerical values, for example: the first level value is 0, the second level value is 1, the third level value is 2, the fourth level value is 3, and the fifth level value is 4.

[0040] Exemplarily, the target user group is a user group of a reading website (for example, all users who have registered on the reading website), and the business data is the reading history content of a user in the target user group. The electronic device can pre-set 5 quality levels, namely: Level 0-poor quality (advertising promotion), Level 1-grayscale (unclear pictures), Level 2-ordinary, Level 3-high quality (written by high-quality authors, exquisite design), Level 4-high-quality (official media release, authoritative value); the electronic device can mark and score the quality of the historical content read by the user through training classification models or manual annotation, thereby obtaining at least one first label information, and each first label information is the identification of the corresponding reading history content, the quality level to which it belongs, and the quality score. It should be noted that other types of business data can also be used in the embodiments of the present application. When classifying the target user group, the same method is used, and they are not listed one by one here.

[0041] Step 201b: The electronic device determines a quality weight set for each quality feature sequence according to usage information and at least one first tag information to obtain a quality feature sequence.

[0042] In an embodiment of the present application, the above-mentioned usage information is the usage information of at least one service data by a user in the target user group, and a quality weight set includes at least one quality weight corresponding to a user, and each quality weight is the quality weight of service data at a quality level.

[0043] In the embodiment of the present application, the above-mentioned one usage information is at least one usage information corresponding to at least one piece of business data.

[0044] Optionally, in an embodiment of the present application, the electronic device may obtain at least one first tag information of at least one piece of business data of a certain user within a time period and usage information of at least one piece of business data and identification information of the user.

[0045] Optionally, in an embodiment of the present application, each piece of first tag information in the at least one first tag information may be an identifier of the business data and a quality score of the business data.

[0046] Optionally, in an embodiment of the present application, the at least one usage information is the time when the electronic device obtains at least one piece of business data and the interval between the historical usage time of at least one piece of business data. The at least one historical usage time of the business data may be the start time or the end time of the use of the business data. For example, the electronic device obtains a piece of business data on March 6, and the user starts using the piece of business data on March 3. At this time, the time interval is 3, and the usage information of the piece of business data is 3.

[0047] Optionally, in the embodiment of the present application, each piece of usage information is represented by t.

[0048] Optionally, in an embodiment of the present application, the user's identification information may include at least one of the following: the user's registered account, the user's nickname, and the user's registered mobile phone number.

[0049] Optionally, in an embodiment of the present application, the electronic device may calculate the quality weight of the service data at each quality level by using the following formula: Q=∑a*b*exp^(-α*t).

[0050] Among them, Q is the quality weight of the business data under each quality level, a is the quality level to which each business data belongs, b is the quality score of each business data, α is the attenuation coefficient, and t is the interval between the time when the electronic device obtains each business data and the last use time / first use time of each business data.

[0051] Exemplarily, taking the target user group as the user group of a reading website (for example, all users who have registered on the reading website) and the business data as the reading history content of a user in the target user group as an example, the electronic device can obtain the reading history content of the user in the past 30 days, and each content record includes: user ID, first tag information (that is, the ID and quality score of the reading history content), usage information (that is, the interval between the time when the electronic device obtains the reading history content and the time when the user starts reading the reading history content, for example: the time when the electronic device obtains the reading history content is March 6, and the time when the user starts reading the reading history content is March 3, then the time interval is 3, and the usage information of the reading history content is 3), then the electronic device can calculate the quality weight of the reading history content at each quality level according to the preset 5 quality levels, thereby obtaining a quality weight set corresponding to the user, and then obtaining a quality feature sequence (Q0, Q1, Q2, Q3, Q4).

[0052] For example, for user A, assuming that the at least one business data includes data 1 to data 10, and the quality levels of the at least one business data are 0 to 4, respectively, wherein the quality level of data 1 and data 3 is 0, the quality level of data 2, data 5 and data 6 is 1, the quality level of data 4 and data 10 is 2, the quality level of data 7 is 3, and the quality level of data 8 and data 9 is 4; then, the quality weight of the data at quality level 0 (i.e., data 1 and data 3) is Q0, Q0 = a0*b0*exp^(-α*t0)+a1*b1*exp^(-α*t1), a0 is the quality level of data 1, b0 is the quality score of data 1, t0 is the interval between the time when the electronic device obtains data 1 and the last time data 1 is used, a1 is the quality level of data 3, b1 is the quality score of data 3, and t1 is the interval between the time when the electronic device obtains data 3 and the last time data 3 is used; the data at quality level 1 (i.e., data 2, data 5 and data 3) According to 6), the quality weight is Q1, Q1 = a2*b2*exp^(-α*t2)+a3*b3*exp^(-α*t3)+a4*b4*exp^(-α*t4), a2 is the quality level of data 2, b2 is the quality score of data 2, t2 is the interval between the time when the electronic device obtains data 2 and the last use time of data 2, a3 is the quality level of data 5, b3 is the quality score of data 5, t3 is the interval between the time when the electronic device obtains data 5 and the last use time of data 5. The interval between the times, a4 is the quality level of data 6, b4 is the quality score of data 6, t4 is the interval between the time when the electronic device obtains data 6 and the last time data 6 was used; the quality weight of data at quality level 2 (i.e. data 4 and data 10) is Q2; the quality weight of data at quality level 3 (i.e. data 7) is Q3; the quality weight of data at quality level 4 (i.e. data 8 and data 9) is Q4; then the quality feature sequence of user A is (Q0, Q1, Q2, Q3, Q4). The calculation methods for Q2, Q3 and Q4 are similar to those for Q0 and Q1, and are not listed here one by one.

[0053] It should be noted that, for each quality feature sequence in the at least one quality feature sequence, it can be obtained by executing the above steps 201a and 201b.

[0054] Optionally, in an embodiment of the present application, "the electronic device constructs at least one original data feature sequence" in the above step 201 can be specifically implemented through the following step 201c.

[0055] Step 201c: The electronic device determines at least one original data feature for each original data feature sequence according to the first usage information, the second usage information, the third usage information and an identifier of at least one piece of business data to obtain an original data feature sequence.

[0056] In an embodiment of the present application, the first usage information is the most recent usage time of at least one piece of business data by a user in the target user group, the second usage information is the total number of days a user has used at least one piece of business data, and the third usage information is the cumulative duration of use of at least one piece of business data by a user.

[0057] In an embodiment of the present application, the electronic device can determine at least one original data feature based on the most recent usage time, total number of days of use, cumulative usage time and identification of at least one business data of each user, thereby obtaining an original data feature sequence.

[0058] Optionally, in an embodiment of the present application, the most recent usage time of the at least one piece of business data is the interval between the time when the electronic device acquires the at least one piece of business data and the historical time when the user last used the piece of business data. The historical time when the user last used the piece of business data may be the start time or end time of the most recent use of the business data. For example, if the electronic device acquires a piece of business data on March 6, and the historical time when the user last started using the piece of business data is March 3, the time interval is 3, and the usage information of the business data is 3.

[0059] Optionally, in an embodiment of the present application, the total number of days of use of the at least one piece of business data is the cumulative number of active days that the user uses the business data within a time period.

[0060] Optionally, in an embodiment of the present application, the cumulative usage duration of the at least one piece of service data is the cumulative usage duration of the service data by the user within a time period.

[0061] Optionally, in an embodiment of the present application, the most recent usage time of the at least one business data (i.e., the first usage information) can be represented by R, the total number of days of use of the at least one business data (i.e., the second usage information) can be represented by F, and the cumulative usage duration of the at least one business data (i.e., the third usage information) can be represented by M.

[0062] Optionally, in an embodiment of the present application, the electronic device can obtain at least one first tag information of at least one piece of business data of a user within a time period and usage information of at least one piece of business data and identification information of the user to calculate each original data feature, where the at least one first tag information is the identification of at least one piece of business data.

[0063] Optionally, in an embodiment of the present application, the at least one usage information is the time when the electronic device obtains at least one piece of business data and the interval between the historical usage time of at least one piece of business data. The at least one historical usage time of the business data may be the start time or the end time of the use of the business data. For example, the electronic device obtains a piece of business data on March 6, and the user starts using the piece of business data on March 3. At this time, the time interval is 3, and the usage information of the piece of business data is 3.

[0064] Optionally, in an embodiment of the present application, the user's identification information may be any one of the following: the user's registered account, the user's nickname, or the user's registered mobile phone number.

[0065] Exemplarily, the electronic device can obtain the user's reading history content for the past 30 days, and each content record includes: user ID, first tag information (i.e., the ID of the reading history content), and usage information (i.e., the interval between the time when the electronic device obtains the reading history content and the time when the user starts reading the reading history content).

[0066] Optionally, in an embodiment of the present application, the electronic device determines at least one original data feature based on the first usage information, the second usage information, the third usage information and the identifier of at least one business data to obtain an original data feature sequence. Specifically, at least one original data feature may be determined based on at least one usage information corresponding to at least one business data and at least one first tag information (i.e., the identifier w of each business data) to obtain an original data feature sequence.

[0067] Optionally, in an embodiment of the present application, the above-mentioned original data feature sequence can be represented by (R, F, M).

[0068] It can be understood that the electronic device can obtain the user's usage information of at least one business data within a time period and the time when the at least one business data is obtained, so that the electronic device can obtain the interval between the time when the at least one business data is obtained and the historical time when the user last used the business data, that is, determine the first usage information (an original data feature R).

[0069] Exemplarily, the electronic device can determine the time when the user last started reading a certain historical content on a reading website (for example, March 9) based on the usage information of the user's reading history content in the past 30 days. The electronic device can then obtain the interval 6 between the time when the user last started reading a certain historical content on the reading website and the time when the electronic device obtained the user's reading history content in the past 30 days (for example, March 15) to obtain the first usage information, thereby determining a raw data feature R.

[0070] It can be understood that the electronic device can obtain the usage information of at least one business data of the user within a time period, and the usage information includes the time when the user uses at least one business data, so that the electronic device can obtain the cumulative active days that the user uses the business data within a time period, that is, determine the second usage information (an original data feature F).

[0071] Exemplarily, the electronic device can determine the time when the user read the historical content on the reading website (for example, March 3, March 5 to March 15) based on the usage information of the user's reading history content in the past 30 days. The electronic device can then obtain the second usage information and determine an original data feature F.

[0072] It can be understood that the electronic device can obtain the usage information of at least one business data of the user within a time period, and the usage information includes the start time and end time of the user's use of at least one business data, so that the electronic device can obtain the cumulative duration of the user's use of the business data within a time period, that is, determine the third usage information (an original data feature M).

[0073] Exemplarily, the electronic device can determine the start time and end time of all historical content that the user has read on the reading website in the past 30 days based on the usage information of the user's reading history content in the past 30 days (for example: the start time for the user to read the first reading history content is 9:00 on March 3, and the end time is 10:00 on March 3; the start time for the user to read the second reading history content is 10:00 on March 3, and the end time is 12:00 on March 3; the start time for the user to read the third reading history content is 10:00 on March 8, and the end time is 12:00 on March 8), so that the electronic device can obtain the user's cumulative active time on the reading website in the past 30 days as 5 hours, to obtain the third usage information, determine a raw data feature M, and then determine a raw data feature sequence (R, F, M).

[0074] It should be noted that, for each original data feature sequence in the at least one original data feature sequence, it can be obtained by executing the above step 201c.

[0075] Step 202: The electronic device classifies the target user group according to at least one quality feature sequence, at least one original data feature sequence, and at least one user's behavior attribute feature group.

[0076] Optionally, in an embodiment of the present application, the behavioral attribute feature group of the above-mentioned at least one user may include at least one of the following: the user's age, the user's gender, the user's educational background, the user's city, the user's income, the user's occupation, the user's personality, the type of business data the user is concerned about, the form in which the user uses business data, the user's preferences, the time period when the user uses business data, the frequency of the user's use of business data, the duration of the user's single use of business data, the user's demands for using business data, the user's behavior in using business data, and the user's concept of using business data.

[0077] For example, Figure 2 As shown, the user's behavioral attribute feature group may include: the user's age, the user's gender, the user's education level, the user's city, the user's income, the user's occupation, the user's personality, the type of business data the user is interested in, the form in which the user uses business data, the user's preferences, the time period in which the user uses business data, the frequency of the user's use of business data, the duration of the user's single use of business data, the user's demands for using business data, the user's behavior in using business data, the user's concept of using business data, and examples of each content.

[0078] Optionally, in the embodiment of the present application, the behavioral attribute feature group of the at least one user may be represented as W.

[0079] In an embodiment of the present application, the electronic device obtains at least one combined feature group RFMQ based on at least one quality feature sequence, at least one original data feature sequence and at least one user's behavioral attribute feature group, and each combined feature group can be represented by (R, F, M, Q0, Q1, Q2, Q3, Q4, W).

[0080] Optionally, in an embodiment of the present application, the electronic device may classify at least one user of the target user group using a Density-Based Spatial Clustering of Applications with Noise (DBSCAN) clustering algorithm according to the at least one combined feature group mentioned above.

[0081] It can be understood that the electronic device inputs the above-mentioned at least one combined feature group as a data set into the DBSCAN clustering algorithm to classify the target user group.

[0082] Optionally, in the embodiment of the present application, the target group can be divided into any number of user groups, for example, 5 user groups, 8 user groups, etc., which can be adjusted according to business needs, and the embodiment of the present application is not limited.

[0083] The embodiment of the present application provides a user classification method, and the electronic device can classify the target user group according to at least one quality feature sequence of the target user group, at least one original data feature sequence, and at least one behavioral attribute feature group of the user. In this scheme, since the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for each user of the target user group, when the electronic device classifies the target user group, it can classify the target user group in combination with the quality of at least one business data, the user's use of at least one business data, and the user's behavioral attribute feature group, that is, the electronic device classifies the user group based on the quality of the business data, the use of the business data, and the user's behavioral attributes, so that the division of the user group is more accurate, thereby improving the accuracy of the electronic device in classifying users.

[0084] Optionally, in the embodiment of the present application, combined with Figure 1 ,like Figure 3 As shown, after the above step 202, the user classification method provided in the embodiment of the present application also includes the following steps 301 to 303.

[0085] Step 301: The electronic device marks each user in at least one user group to obtain at least one second tag information.

[0086] In an embodiment of the present application, at least one user group is a user group obtained after classifying the target user group, and each second label information includes the total number of days a user uses at least one business data, the cumulative duration of use of at least one business data by a user, and the sum of at least one quality weight corresponding to a user, and each quality weight is the quality weight of business data at a quality level.

[0087] In an embodiment of the present application, the electronic device can mark each user in at least one user group to obtain at least one second label information, so that the electronic device can determine at least one set of average values ​​based on the at least one second label information, and thereby determine the attribute information of at least one user group based on the at least one set of average values.

[0088] Optionally, in an embodiment of the present application, the total number of days that the above-mentioned user uses at least one piece of business data can be expressed in days (for example, 3 days) or in specific hours (for example, 72 hours).

[0089] Optionally, in the embodiment of the present application, the total number of days that the user uses at least one piece of service data is represented by an RF value. Specifically, RF=λR+μF, where λ and μ are hyperparameters.

[0090] Optionally, in the embodiment of the present application, the cumulative duration of use of at least one piece of service data by the above-mentioned one user can be represented by an M value.

[0091] In an embodiment of the present application, the sum of at least one quality weight corresponding to a user is the sum of the quality weights of business data at at least one quality level. For example, a user has quality weights for business data at the first quality level, the second quality level, the third quality level, and the fifth quality level. Then, the sum of at least one quality weight corresponding to the user is the sum of the quality weights of the business data at the first quality level, the second quality level, the third quality level, and the fifth quality level.

[0092] Optionally, in an embodiment of the present application, the sum of at least one quality weight corresponding to a user can be represented by Q.

[0093] In the embodiment of the present application, the electronic device marks each user, and the obtained at least one second tag information also includes user group information to which each user belongs.

[0094] Optionally, in an embodiment of the present application, the user group information to which each user belongs can be represented by U, and the second label information of each user can be represented by (U, RF, M, Q).

[0095] It should be noted that for the description of the total number of days a user uses at least one piece of business data, the cumulative duration of a user's use of at least one piece of business data, and the quality weight of business data at at least one quality level, please refer to the description in the above step 201c, which will not be repeated here.

[0096] Step 302: The electronic device determines at least one average value set according to at least one second tag information.

[0097] In an embodiment of the present application, each set of average values ​​includes a first average value, a second average value and a third average value, the first average value being the average value of the total number of days that all users in a user group use at least one service data, the second average value being the average value of the cumulative duration of use of at least one service data by all users in a user group, and the third average value being the average value of the sum of at least one quality weight corresponding to all users in a user group.

[0098] In an embodiment of the present application, the at least one average value set is a set of average values ​​corresponding to at least one user group, the first average value in each average value set is the average of the total number of days that all users in the corresponding user group use at least one business data, the second average value in each average value set is the average of the cumulative duration of use of at least one business data by all users in the corresponding user group, and the third average value in each average value set is the average of the sum of at least one quality weight corresponding to all users in the corresponding user group.

[0099] Specifically, the set formula for calculating the average value of each user group is: RF_ave=∑RF / total number of users in the group, M_ave=∑M / total number of users in the group, Q_ave=∑Q / total number of users in the group.

[0100] Among them, RF is the total number of days that a user uses at least one service data, M is the cumulative duration of use of at least one service data by a user, and Q is the sum of at least one quality weight corresponding to a user.

[0101] For example, for user group A, assuming that user group A includes user 1, user 2, and user 3, the first average value RF1_ave in the average value set corresponding to user group A = RF1+RF2+RF3 / total number of users in the group 3, RF1 is the total number of days that user 1 uses at least one service data, RF2 is the total number of days that user 2 uses at least one service data, and RF3 is the total number of days that user 3 uses at least one service data; the second average value M1_ave in the average value set corresponding to user group A = M1+M2+M3 / total number of users in the group 3, M1 is the total number of days that user 1 uses at least one service data The third average value in the average value set corresponding to user group A is Q1_ave=Q1+Q2+Q3 / total number of users in the group 3, Q1 is the sum of at least one quality weight corresponding to user 1, Q2 is the sum of at least one quality weight corresponding to user 2, and Q3 is the sum of at least one quality weight corresponding to user 3; then the average value set corresponding to user group A is (RF1_ave, M1_ave, Q1_ave). The calculation method for other user groups is similar to that for user group A, and is not listed here one by one.

[0102] Step 303: The electronic device determines attribute information of at least one user group according to at least one set of average values.

[0103] Optionally, in an embodiment of the present application, the electronic device compares at least one set of average values ​​corresponding to at least one user group with a threshold preset by the electronic device, thereby determining attribute information of at least one user group.

[0104] Optionally, in the embodiment of the present application, the thresholds preset by the electronic device may be a first threshold RF_threshold, a second threshold M_threshold, and a third threshold Q_threshold.

[0105] Optionally, in an embodiment of the present application, the attribute information of the at least one user grouping includes: potential attribute, active attribute, and taste attribute.

[0106] Optionally, in the embodiment of the present application, for each user group, the attribute information of a user group includes a potential value of a user group in the first dimension attribute, an activity value of a user group in the second dimension attribute, and a taste value of a user group in the third dimension attribute. The above step 303 can be specifically implemented by the following step 303a.

[0107] Step 303a: For each user group, the electronic device determines a potential value of the user group in the first dimension attribute according to a first average value of the user group, determines an activity value of the user group in the second dimension attribute according to a second average value of the user group, and determines a taste value of the user group in the third dimension attribute according to a third average value of the user group.

[0108] Optionally, in an embodiment of the present application, the above-mentioned first dimension attribute is a potential attribute of the user grouping, and the potential value of a user grouping in the first dimension attribute may be high potential / low potential; the second dimension attribute is an active attribute of the user grouping, and the active value of a user grouping in the second dimension attribute may be high active / low active; the above-mentioned third dimension attribute is a quality attribute of the user grouping, and the taste value of a user grouping in the third dimension attribute may be high taste / ordinary.

[0109] Optionally, in the embodiment of the present application, if the first average value of the target user group is greater than or equal to the first threshold, the second average value of the target user group is greater than or equal to the second threshold, and the third average value of the target user group is greater than or equal to the third threshold, that is, RF_ave>=RF_threshol, M_ave>=RF_threshold, Q_ave>=Q_threshold, then the target user group is determined to be a highly active, high-potential, and high-quality user group;

[0110] If the first average value of the target user group is greater than or equal to the first threshold, and the second average value of the target user group is greater than or equal to the second threshold, and the third average value of the target user group is less than the third threshold, that is, RF_ave >= RF_threshold, M_ave >= RF_threshold, Q_ave < Q_threshold, then it is determined that the target user group is a high-activity, high-potential, ordinary user group;

[0111] If the first average value of the target user group is greater than or equal to the first threshold, and the second average value of the target user group is less than the second threshold, and the third average value of the target user group is greater than or equal to the third threshold, that is, RF_ave >= RF_threshold, M_ave < RF_threshold, Q_ave >= Q_threshold, then it is determined that the target user group is a high-activity, low-potential, high-quality user group;

[0112] If the first average value of the target user group is greater than or equal to the first threshold, and the second average value of the target user group is less than the second threshold, and the third average value of the target user group is less than the third threshold, that is, RF_ave >= RF_threshold, M_ave < RF_threshold, Q_ave < Q_threshold, then it is determined that the target user group is a high-activity, low-potential, ordinary user group;

[0113] If the first average value of the target user group is less than the first threshold, and the second average value of the target user group is greater than or equal to the second threshold, and the third average value of the target user group is greater than or equal to the third threshold, that is, RF_ave < RF_threshold, M_ave >= RF_threshold, Q_ave >= Q_threshold, then it is determined that the target user group is a low-activity, high-potential, high-quality user group;

[0114] If the first average value of the target user group is less than the first threshold, and the second average value of the target user group is greater than or equal to the second threshold, and the third average value of the target user group is less than the third threshold, that is, RF_ave < RF_threshold, M_ave >= RF_threshold, Q_ave < Q_threshold, then it is determined that the target user group is a low-activity, high-potential, ordinary user group;

[0115] If the first average value of the target user group is less than the first threshold, the second average value of the target user group is less than the second threshold, and the third average value of the target user group is greater than or equal to the third threshold, that is, RF_ave<RF_threshol, M_ave<RF_threshold, Q_ave>=Q_threshold, then the target user group is determined to be a low-activity, low-potential, high-quality user group;

[0116] If the first average value of the target user group is less than the first threshold, the second average value of the target user group is less than the second threshold, and the third average value of the target user group is less than the third threshold, that is, RF_ave<RF_threshol, M_ave<RF_threshold, Q_ave<Q_threshold, then the target user group is determined to be a low-activity and low-potential ordinary user group.

[0117] For example, Figure 4 As shown, 8 different types of user groups are displayed, as well as the number of users corresponding to each user group and the proportion of the user group in all user groups. For example, the number of users in the highly active, high-potential, and high-quality user group is 12,000, accounting for 12% of all user groups; the number of users in the highly active, high-potential, and ordinary user group is 21,000, accounting for 21% of all user groups.

[0118] Optionally, in the embodiment of the present application, combined with Figure 3 ,like Figure 5 As shown, after the above step 303, the user classification method provided in the embodiment of the present application further includes the following step 401.

[0119] Step 401: The electronic device recommends service content corresponding to the attribute information of each user group for each user group based on the attribute information of each user group.

[0120] In the embodiment of the present application, the electronic device recommends corresponding business content to each user group according to the different attribute information of each user group. For example: for the high-potential, low-active and high-taste group, focus on increasing the consumption time and push high-quality goods, so that the high-potential, low-active and high-taste group can develop into the high-potential, high-active and high-taste group; for the low-potential, high-active ordinary group, increase the user's consumption time and the number of days the user is active, and gradually increase high-quality goods on the basis of the original pushed goods, improve the user's quality attributes, so that the low-potential, high-active ordinary group can develop into the high-potential, high-active and high-taste group; for the low-potential, low-active ordinary group, increase the user's consumption time, the number of days the user is active and the consumption frequency, so that the low-potential, low-active ordinary group can develop into the high-potential, high-active and high-taste group.

[0121] It should be noted that the user classification method provided in the embodiment of the present application can be executed by an electronic device, a user classification device, or a control module in the user classification device. In the embodiment of the present application, the user classification device provided in the embodiment of the present application is described by taking an electronic device executing the user classification method as an example.

[0122] Figure 6 A possible structural diagram of a user classification device involved in an embodiment of the present application is shown. Figure 6 As shown, the user classification device 70 may include: a construction module 71 and a processing module 72 .

[0123] Among them, the construction module 71 is used to construct at least one quality feature sequence and at least one original data feature sequence of the target user group, the target user group includes at least one user, each user corresponds to a quality feature sequence and an original data feature sequence, the quality feature sequence is used to indicate the quality of at least one business data, and the original data feature sequence is used to indicate the user's use of at least one business data. The processing module 72 is used to classify the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavior attribute feature group.

[0124] An embodiment of the present application provides a user classification device. Since the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for each user of the target user group, when the electronic device classifies the target user group, it can classify the target user group based on the quality of at least one business data, the user's use of at least one business data, and the user's behavioral attribute feature group. That is, the electronic device classifies the user group based on the quality of the business data, the use of the business data, and the user's behavioral attributes, which makes the division of user groups more accurate, thereby improving the accuracy of the electronic device in classifying users.

[0125] In a possible implementation, the above-mentioned construction module 71 is specifically used to mark and score at least one business data to obtain at least one first label information, each first label information includes an identifier of a business data, a quality level to which a business data belongs, and a quality score of the business data; and for each quality feature sequence, a quality weight set is determined based on a usage information and at least one first label information to obtain a quality feature sequence, a usage information is the usage information of at least one business data by a user in the target user group, a quality weight set includes at least one quality weight corresponding to a user, and each quality weight is the quality weight of business data at a quality level.

[0126] In a possible implementation, the above-mentioned construction module 71 is specifically used to determine at least one original data feature for each original data feature sequence based on the first usage information, the second usage information, the third usage information and the identifier of at least one business data to obtain an original data feature sequence, wherein the first usage information is the most recent usage time of at least one business data by a user in the target user group, the second usage information is the total number of days that a user uses at least one business data, and the third usage information is the cumulative duration of use of at least one business data by a user.

[0127] In a possible implementation, the user classification device 70 provided in the embodiment of the present application further includes: a marking module and a determination module. The marking module is used to mark each user in at least one user group after the processing module classifies the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavior attribute feature group, and obtain at least one second label information. The at least one user group is a user group obtained after the target user group is classified. Each second label information includes the total number of days a user uses at least one service data, the cumulative duration of use of at least one service data by a user, and the sum of at least one quality weight corresponding to a user, and each quality weight is the quality weight of the service data under a quality level. The determination module is used to determine at least one average value set according to at least one second label information, and each average value set includes a first average value, a second average value and a third average value. A first average value is the average value of the total number of days all users in a user group use at least one service data, a second average value is the average value of the cumulative duration of use of at least one service data by all users in a user group, and a third average value is the average value of the sum of at least one quality weight corresponding to all users in a user group. The determination module is further used to determine attribute information of at least one user group based on at least one average value set.

[0128] In a possible implementation, for each user group, the attribute information of a user group includes a potential value of a user group in a first dimension attribute, an activity value of a user group in a second dimension attribute, and a taste value of a user group in a third dimension attribute; the above-mentioned determination module is specifically used to determine, for each user group, the potential value of a user group in the first dimension attribute according to a first average value of a user group, determine the activity value of a user group in the second dimension attribute according to a second average value of a user group, and determine the taste value of a user group in the third dimension attribute according to a third average value of a user group.

[0129] In a possible implementation, the user classification device 70 provided in the embodiment of the present application further includes: a recommendation module. The recommendation module is used to recommend business content corresponding to the attribute information of each user group for each user group according to the attribute information of each user group after the determination module determines the attribute information of at least one user group according to at least one average value set.

[0130] The user classification device in the embodiment of the present application can be a device, or a component, integrated circuit, or chip in an electronic device. The device can be a mobile electronic device or a non-mobile electronic device. Exemplarily, the mobile electronic device can be a mobile phone, a tablet computer, a laptop computer, a PDA, a car-mounted electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR) / virtual reality (virtual reality, VR) device, a robot, a wearable device, an ultra-mobile personal computer (ultra-mobile personal computer, UMPC), a netbook or a personal digital assistant (personal digital assistant, PDA), etc. It can also be a server, a network attached storage (Network Attached Storage, NAS), a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., and the embodiment of the present application is not specifically limited.

[0131] The user classification device in the embodiment of the present application may be a device having an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in the embodiment of the present application.

[0132] The user classification device provided in the embodiment of the present application can implement each process implemented in the above method embodiment, and will not be described again here to avoid repetition.

[0133] Alternatively, if Figure 7 As shown, an embodiment of the present application also provides an electronic device 900, including a processor 901 and a memory 902, wherein the memory 902 stores programs or instructions that can be executed on the processor 901, and when the program or instructions are executed by the processor 901, the various steps of the above-mentioned user classification method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, they are not repeated here.

[0134] It should be noted that the electronic devices in the embodiments of the present application include the mobile electronic devices and non-mobile electronic devices mentioned above.

[0135] Figure 8 A schematic diagram of the hardware structure of an electronic device to implement an embodiment of the present application.

[0136] The electronic device 100 includes but is not limited to components such as a radio frequency unit 101, a network module 102, an audio output unit 103, an input unit 104, a sensor 105, a display unit 106, a user input unit 107, an interface unit 108, a memory 109, and a processor 110.

[0137] Those skilled in the art will appreciate that the electronic device 100 may also include a power source (such as a battery) for supplying power to various components, and the power source may be logically connected to the processor 110 through a power management system, thereby implementing functions such as managing charging, discharging, and power consumption management through the power management system. Figure 8 The electronic device structure shown in the figure does not constitute a limitation on the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently, which will not be described in detail here.

[0138] The processor 110 is used to construct at least one quality feature sequence and at least one original data feature sequence of a target user group, wherein the target user group includes at least one user, and each user corresponds to a quality feature sequence and an original data feature sequence, wherein the quality feature sequence is used to indicate the quality of at least one piece of business data, and the original data feature sequence is used to indicate the user's use of at least one piece of business data. The processor 110 is used to classify the target user group according to the at least one quality feature sequence, the at least one original data feature sequence, and the at least one user's behavior attribute feature group.

[0139] An embodiment of the present application provides an electronic device. Since the electronic device can construct a corresponding quality feature sequence and a corresponding original data feature sequence for each user of the target user group, when the electronic device classifies the target user group, it can classify the target user group based on the quality of at least one business data, the user's use of at least one business data, and the user's behavioral attribute feature group. That is, the electronic device classifies the user group based on the quality of the business data, the use of the business data, and the user's behavioral attributes, which makes the division of the user group more accurate, thereby improving the accuracy of the electronic device in classifying users.

[0140] Optionally, the processor 110 is specifically used to mark and score at least one piece of business data to obtain at least one first label information, each first label information including an identifier of a piece of business data, a quality level to which a piece of business data belongs, and a quality score of the business data; for each quality feature sequence, a quality weight set is determined based on a usage information and at least one first label information to obtain a quality feature sequence, a usage information is usage information of at least one piece of business data by a user in a target user group, a quality weight set includes at least one quality weight corresponding to a user, and each quality weight is a quality weight of business data at a quality level.

[0141] Optionally, the processor 110 is specifically used to determine at least one original data feature for each original data feature sequence based on first usage information, second usage information, third usage information and an identifier of at least one business data to obtain an original data feature sequence, wherein the first usage information is the most recent usage time of at least one business data by a user in the target user group, the second usage information is the total number of days that a user has used at least one business data, and the third usage information is the cumulative duration of use of at least one business data by a user.

[0142] Optionally, the processor 110 is further used to classify the target user group according to at least one quality feature sequence, at least one original data feature sequence and at least one user's behavior attribute feature group, and then mark each user in at least one user group to obtain at least one second tag information, at least one user group is a user group obtained after classifying the target user group, and each second tag information includes the total number of days a user uses at least one business data, the cumulative duration of use of at least one business data by a user, and the sum of at least one quality weight corresponding to a user, and each quality weight is the quality weight of the business data under a quality level. And according to at least one second tag information, at least one average value set is determined, each average value set includes a first average value, a second average value and a third average value, a first average value is the average value of the total number of days all users in a user group use at least one business data, a second average value is the average value of the cumulative duration of use of at least one business data by all users in a user group, and a third average value is the average value of the sum of at least one quality weight corresponding to all users in a user group. And according to at least one average value set, the attribute information of at least one user group is determined.

[0143] Optionally, for each user group, the attribute information of a user group includes a potential value of a user group in a first dimension attribute, an activity value of a user group in a second dimension attribute, and a taste value of a user group in a third dimension attribute; the processor 110 is specifically used to determine, for each user group, a potential value of a user group in the first dimension attribute according to a first average value of the user group, determine an activity value of a user group in the second dimension attribute according to a second average value of the user group, and determine a taste value of a user group in the third dimension attribute according to a third average value of the user group.

[0144] Optionally, the processor 110 is further configured to, after determining attribute information of at least one user group based on at least one set of average values, recommend business content corresponding to the attribute information of each user group for each user group based on the attribute information of each user group.

[0145] The electronic device provided in the embodiment of the present application can implement each process implemented in the above method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described here.

[0146] The beneficial effects of various implementations in this embodiment can be specifically referred to the beneficial effects of the corresponding implementations in the above method embodiment. To avoid repetition, they will not be described again here.

[0147] It should be understood that in the embodiment of the present application, the input unit 104 may include a graphics processor (Graphics Processing Unit, GPU) 1041 and a microphone 1042, and the graphics processor 1041 processes the image data of a static picture or video obtained by an image capture device (such as a camera) in a video capture mode or an image capture mode. The display unit 106 may include a display panel 1061, and the display panel 1061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc. The user input unit 107 includes a touch panel 1071 and at least one of other input devices 1072. The touch panel 1071 is also called a touch screen. The touch panel 1071 may include two parts: a touch detection device and a touch controller. Other input devices 1072 may include, but are not limited to, a physical keyboard, function keys (such as volume control keys, switch keys, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.

[0148] The memory 109 can be used to store software programs and various data. The memory 109 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc. In addition, the memory 109 may include a volatile memory or a non-volatile memory, or the memory 109 may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM). The memory 109 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

[0149] The processor 110 may include one or more processing units; optionally, the processor 110 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 110.

[0150] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0151] The processor is the processor in the electronic device described in the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.

[0152] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0153] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0154] An embodiment of the present application provides a computer program product, which is stored in a storage medium. The program product is executed by at least one processor to implement the various processes of the above-mentioned user classification method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0155] It should be noted that, in this article, the terms "comprise", "include" or any other variant thereof are intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise one..." do not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0156] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM / RAM, a disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0157] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A user classification method, characterized in that: The method comprises: Constructing at least one quality feature sequence and at least one original data feature sequence of a target user group, wherein the target user group includes at least one user, and each user corresponds to a quality feature sequence and an original data feature sequence, wherein the quality feature sequence is used to indicate the quality of at least one piece of service data, and the original data feature sequence is used to indicate the user's use of the at least one piece of service data; Classifying the target user group according to the at least one quality feature sequence, the at least one original data feature sequence and the at least one user's behavior attribute feature group; The step of constructing at least one quality feature sequence of the target user group includes: Marking and scoring the at least one piece of service data to obtain at least one piece of first label information, each piece of first label information including an identifier of a piece of service data, a quality level to which the piece of service data belongs, and a quality score of the piece of service data; For each quality feature sequence, a quality weight set is determined according to usage information and the at least one first tag information to obtain a quality feature sequence, wherein the usage information is usage information of the at least one service data by a user in the target user group, the quality weight set includes at least one quality weight corresponding to the user, and each quality weight is a quality weight of service data at a quality level; The constructing of at least one original data feature sequence comprises: For each original data feature sequence, at least one original data feature is determined based on first usage information, second usage information, third usage information and the identifier of the at least one business data to obtain an original data feature sequence, wherein the first usage information is the most recent usage time of the at least one business data by a user in the target user group, the second usage information is the total number of days that the at least one business data is used by the user, and the third usage information is the cumulative duration of use of the at least one business data by the user.

2. The method according to claim 1, characterized in that After classifying the target user group according to the at least one quality feature sequence, the at least one original data feature sequence and the at least one user's behavior attribute feature group, the method further includes: Marking each user in at least one user group to obtain at least one second label information, wherein the at least one user group is a user group obtained by classifying the target user group, and each second label information includes a total number of days that a user uses the at least one service data, a cumulative duration of use of the at least one service data by the user, and a sum of at least one quality weight corresponding to the user, and each quality weight is a quality weight of service data at a quality level; Determine at least one set of average values ​​according to the at least one second tag information, each set of average values ​​includes a first average value, a second average value, and a third average value, the first average value is an average value of the total number of days that all users in a user group use the at least one service data, the second average value is an average value of the cumulative duration of use of the at least one service data by all users in a user group, and the third average value is an average value of the sum of at least one quality weight corresponding to all users in a user group; According to the at least one set of average values, attribute information of the at least one user group is determined.

3. The method according to claim 2, characterized in that For each user group, the attribute information of a user group includes a potential value of the user group in the first dimension attribute, an activity value of the user group in the second dimension attribute, and a taste value of the user group in the third dimension attribute; The determining, according to the at least one average value set, the attribute information of the at least one user grouping includes: For each user group, the potential value of the user group in the first dimension attribute is determined based on the first average value of the user group, the activity value of the user group in the second dimension attribute is determined based on the second average value of the user group, and the taste value of the user group in the third dimension attribute is determined based on the third average value of the user group.

4. A user classification device, characterized in that: The user classification device comprises: a construction module and a processing module; The construction module is used to construct at least one quality feature sequence and at least one original data feature sequence of a target user group, wherein the target user group includes at least one user, and each user corresponds to a quality feature sequence and an original data feature sequence, wherein the quality feature sequence is used to indicate the quality of at least one piece of service data, and the original data feature sequence is used to indicate the user's use of the at least one piece of service data; The processing module is used to classify the target user group according to the at least one quality feature sequence, the at least one original data feature sequence and the at least one user's behavior attribute feature group; The construction module is specifically used to mark and score the at least one piece of business data to obtain at least one first label information, each of which includes an identifier of a piece of business data, a quality level to which the piece of business data belongs, and a quality score of the piece of business data; and for each quality feature sequence, determine a quality weight set according to a usage information and the at least one first label information to obtain a quality feature sequence, the usage information is usage information of the at least one piece of business data by a user in the target user group, the quality weight set includes at least one quality weight corresponding to the user, and each quality weight is a quality weight of business data at a quality level; The construction module is specifically used to determine at least one original data feature for each original data feature sequence based on first usage information, second usage information, third usage information and an identifier of the at least one business data to obtain an original data feature sequence, wherein the first usage information is the most recent usage time of the at least one business data by a user in the target user group, the second usage information is the total number of days that the at least one business data is used by the user, and the third usage information is the cumulative duration of use of the at least one business data by the user.

5. The device according to claim 4, characterized in that The user classification device further includes: a marking module and a determination module; The marking module is used to mark each user in at least one user group after the processing module classifies the target user group according to the at least one quality feature sequence, the at least one original data feature sequence and the behavioral attribute feature group of the at least one user, so as to obtain at least one second label information, wherein the at least one user group is a user group obtained after classifying the target user group, and each second label information includes the total number of days that a user uses the at least one service data, the cumulative duration of use of the at least one service data by the user, and the sum of at least one quality weight corresponding to the user, and each quality weight is a quality weight of service data at a quality level; The determination module is used to determine at least one set of average values ​​based on the at least one second tag information, each set of average values ​​includes a first average value, a second average value and a third average value, the first average value is the average value of the total number of days that all users in a user group use the at least one business data, the second average value is the average value of the cumulative duration of use of the at least one business data by all users in a user group, and the third average value is the average value of the sum of at least one quality weight corresponding to all users in a user group; and determine the attribute information of the at least one user group based on the at least one set of average values.

6. The device according to claim 5, characterized in that For each user group, the attribute information of a user group includes a potential value of the user group in the first dimension attribute, an activity value of the user group in the second dimension attribute, and a taste value of the user group in the third dimension attribute; The determination module is specifically used to determine, for each user group, the potential value of the user group in the first dimensional attribute according to the first average value of the user group, determine the activity value of the user group in the second dimensional attribute according to the second average value of the user group, and determine the taste value of the user group in the third dimensional attribute according to the third average value of the user group.

7. An electronic device, characterized in that: The method comprises a processor, a memory and a program or instruction stored in the memory and executable on the processor, wherein the program or instruction, when executed by the processor, implements the steps of the user classification method as claimed in any one of claims 1 to 3.

8. A readable storage medium, characterized in that: The readable storage medium stores a program or an instruction, and when the program or the instruction is executed by a processor, the steps of the user classification method according to any one of claims 1 to 3 are implemented.

Images