User classification method and related device

Through the analysis and classification technology of user browsing behavioral data, the problem of marketers being difficult to accurately determine user preferences is solved, the users' precise classification and targeted marketing activities are achieved, and marketing efficiency is improved.

CN112001761BActive Publication Date: 2024-11-08SHANGHAI PATEO ELECTRONIC EQUIPMENT MANUFACTURING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010896133.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2020-08-31
Publication Date
2024-11-08
Estimated Expiration
2040-08-31

Smart Images

  • Figure CN112001761B_ABST
    Figure CN112001761B_ABST
Patent Text Reader

Abstract

The embodiment of the present application discloses a user classification method and related devices, the method includes: obtaining browsing behavior data of each user, the browsing behavior data of the first user includes: the browsing times of the first user for M categories; determining the first similarity between the first user and each classification benchmark user according to the browsing times of the first user for M categories; determining the classification benchmark user matching the first user according to the first similarity corresponding to the first user and each classification benchmark user; classifying the first user into the reference group corresponding to the matching classification benchmark user; determining the similarity between the first user and the matching classification benchmark user in each category according to the browsing times of the first user for each category in the M categories, obtaining M second similarities corresponding to the first user and the matching classification benchmark user; classifying the first user into the target group corresponding to the M second similarities. The embodiment of the present application is conducive to accurate classification of users.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a user classification method and related devices. Background Art

[0002] In the arrangement of various marketing business activities, such as vehicle marketing, when exploring potential customers, marketers often need to understand the preferences and needs of different customers in order to arrange marketing activities in a targeted manner. Since user preferences are often complex and comprehensive, marketers cannot accurately determine the preferences of each user for a large number of users, and cannot conduct differentiated marketing interactions for different types of users, resulting in low marketing efficiency. Summary of the invention

[0003] The embodiments of the present application provide a user classification method and related devices in order to accurately classify users.

[0004] In a first aspect, an embodiment of the present application provides a user classification method, including:

[0005] Obtain browsing behavior data of each user of at least one user, where the browsing behavior data of a first user includes: browsing times of the first user for M categories, where M is a positive integer, and the first user is any one of the at least one user;

[0006] Determining a first similarity between the first user and each classification reference user according to the browsing times of the first user for the M categories;

[0007] Determining a classification reference user matching the first user according to a first similarity between the first user and each classification reference user;

[0008] Classifying the first user into a reference group corresponding to the matched classification reference user, wherein the reference group corresponds to a plurality of target groups;

[0009] Determine the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user for each of the M categories, and obtain M second similarities corresponding to the first user and the matched classification reference user;

[0010] The first user is divided into a target group corresponding to the M second similarities among the multiple target groups.

[0011] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, determining the first similarity between the first user and each classification reference user according to the number of browsing times of the first user for the M categories includes:

[0012] Establish a multi-dimensional coordinate system;

[0013] Determining a first feature vector of the first user in the multi-dimensional coordinate system according to the browsing times of the M categories;

[0014] Determine a second feature vector of each classification reference user in the multi-dimensional coordinate system;

[0015] The cosine similarity between the first feature vector and the second feature vector corresponding to each classification reference user is calculated, and the cosine similarity is used as the first similarity between the first user and each classification reference user.

[0016] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, the dividing the first user into the target groups corresponding to the M second similarities among the multiple target groups includes:

[0017] Determining a first classification standard for the first user according to the M second similarities corresponding to the first user;

[0018] Determine a first preset classification standard that matches the first classification standard from a plurality of preset classification standards, wherein the plurality of preset classification standards correspond one-to-one to the plurality of target groups;

[0019] The first user is divided into a target group corresponding to the first preset classification standard.

[0020] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, when the classification reference user is a user who has purchased a car, the method further includes:

[0021] Acquire vehicle parameters of the vehicle purchased by the classification reference user, wherein the vehicle parameters include at least one of the following: vehicle model, color, size, performance, and price;

[0022] Acquire a candidate vehicle that matches vehicle parameters of the classification reference user's purchased vehicle from a database, wherein the database stores data on vehicles currently on sale by a car seller;

[0023] The candidate vehicle is pushed to the first user.

[0024] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, when the classification reference user is a user who has purchased a financial product, the method further includes:

[0025] Acquire product parameters of the wealth management products purchased by the classification basis user, wherein the product parameters include at least one of the following: product category, risk level, profitability, and liquidity of funds;

[0026] Acquire, from a wealth management product database, candidate wealth management products that match product parameters of wealth management products purchased by the classification basis user;

[0027] The candidate financial product is pushed to the first user.

[0028] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, when the classification reference user is a user who has purchased vehicle insurance, the method further includes:

[0029] Obtaining insurance information of vehicle insurance purchased by the classification base user, wherein the insurance information includes at least one of the following: insurance type, insurance cost, coverage amount, insurance company, and insurance company credit rating;

[0030] Acquire, from a vehicle insurance database, candidate vehicle insurance that matches insurance information of vehicle insurance purchased by the classification reference user;

[0031] The candidate vehicle insurance is pushed to the first user.

[0032] In combination with the first aspect of the present application, in a possible implementation of the first aspect of the present application, when the classification reference user is a user who has purchased travel services, the method further includes:

[0033] Acquire service information of the travel service purchased by the classification base user, wherein the service information includes at least one of the following: travel destination, travel mode, travel duration, and travel cost composition;

[0034] Acquire, from a travel service database, candidate travel services that match the service information of the travel services purchased by the classification reference user;

[0035] The candidate travel service is pushed to the first user.

[0036] In a second aspect, an embodiment of the present application provides a user classification device, including:

[0037] an acquisition unit, configured to acquire browsing behavior data of each user of at least one user, wherein the browsing behavior data of a first user includes: browsing times of the first user for M categories, wherein M is a positive integer, and the first user is any one of the at least one user;

[0038] A first determining unit, configured to determine a first similarity between the first user and each classification reference user according to the number of views of the first user on the M categories;

[0039] a second determining unit, configured to determine a classification reference user matching the first user according to a first similarity between the first user and each classification reference user;

[0040] A first grouping unit, configured to classify the first user into a reference group corresponding to the matched classification reference user, wherein the reference group corresponds to a plurality of target groups;

[0041] A third determining unit is used to determine the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user for each of the M categories, and obtain M second similarities corresponding to the first user and the matched classification reference user;

[0042] The second grouping unit is configured to group the first user into a target group corresponding to the M second similarities among the multiple target groups.

[0043] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, and a communication interface, wherein the processor, the memory, and the communication interface are interconnected, wherein the communication interface is used to receive or send data, the memory is used to store application code for the electronic device to execute the above method, and the processor is configured to execute any method of the above first aspect.

[0044] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, a computer program for electronic data exchange, wherein the above-mentioned computer program enables a computer to execute part or all of the steps described in any method of the first aspect of the embodiment of the present application.

[0045] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in any method of the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0046] It can be seen that in the embodiment of the present application, the electronic device first obtains the browsing behavior data of each user, and the browsing behavior data of the first user includes: the number of times the first user browses M categories, and then determines the first similarity between the first user and each classification benchmark user based on the number of times the first user browses the M categories, and further divides the first user into a reference group corresponding to the matching classification benchmark user based on the first similarity, and then determines the similarity between the first user and the matching classification benchmark user in each category based on the number of times the first user browses each of the M categories, and divides the first user into a target group corresponding to the M second similarities. It can be seen that according to the number of times the user browses the M categories, the first similarity and M second similarities between the user and the classification benchmark user are determined respectively, and the user is grouped twice based on the first similarity and the second similarity, which is conducive to accurate classification of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0048] Figure 1 This is an architecture diagram of a user classification system provided in an embodiment of the present application;

[0049] Figure 2 It is a flowchart of a user classification method provided in an embodiment of the present application;

[0050] Figure 3 It is a schematic diagram of the structure of a user classification device provided in an embodiment of the present application;

[0051] Figure 4 It is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only 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.

[0053] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0054] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0055] "Multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship.

[0056] The embodiments of the present application are described below in conjunction with the accompanying drawings.

[0057] See also Figure 1 , Figure 1 This is an architecture diagram of a user classification system provided in an embodiment of the present application. As shown in the figure, the user classification system may include a server 100 and a terminal device 101. The terminal device 101 may be one or more. Each terminal device 101 may establish a communication connection with the server. The server 100 may obtain the browsing behavior data of the user sent from the terminal device 101, and may also send corresponding data to the terminal device 101. In addition, the server 100 may also establish a communication connection with other servers to obtain data stored in databases of other servers, such as obtaining vehicle parameters stored in a vehicle database, or a corresponding database such as a vehicle parameter database is set in the server 100.

[0058] See also Figure 2 , Figure 2 : is a flowchart of a user classification method provided in an embodiment of the present application. As shown in the figure, the user classification method includes the following operations:

[0059] S201, the electronic device obtains browsing behavior data of each user of at least one user, where the browsing behavior data of a first user includes: the number of browsing times of M categories by the first user.

[0060] Wherein, M is a positive integer, the first user is any one of the at least one user; the electronic device can be as follows Figure 1 The server shown.

[0061] In the specific implementation, the user's browsing behavior data can be collected by data embedding. The browsing behavior data can be generated by the user's online browsing behavior, such as the data generated by the user browsing in various public accounts, applets or mobile applications. Considering the different needs for classifying users in different scenarios, M categories can be pre-set according to actual needs. When obtaining the user's browsing behavior data, only the user's browsing behavior data for the pre-set M categories is obtained. Of course, all browsing behavior data of each user can also be obtained, and the M categories include each category browsed by each user.

[0062] In addition, the classification of categories can also be adjusted according to actual conditions. In different scenarios, different content can be classified into different categories. For example, in the scenario of mining potential customers for freight trucks, various content related to family cars can be classified into one category. When mining potential customers for a certain car model suitable for family use, the content related to family cars can be divided into multiple categories, such as content related to small family cars is classified into one category, and content related to large family cars is classified into another category.

[0063] S202: The electronic device determines a first similarity between the first user and each classification reference user according to the number of views of the first user on the M categories.

[0064] S203: The electronic device determines a classification reference user that matches the first user according to a first similarity between the first user and each classification reference user.

[0065] S204: The electronic device divides the first user into a reference group corresponding to the matched classification reference user, where the reference group corresponds to a plurality of target groups.

[0066] In S202, S203 and S204, the classification reference user can be a virtual user set according to actual needs, or a specific user selected; the data of the classification reference user can be one or more, which can be set according to actual needs and is not limited here. For a user, the user's preferences for different categories can be understood by combining the number of views of the user for multiple categories. Therefore, the first similarity between the first user and each classification reference user determined based on the number of views of the first user for M categories can be used to characterize the similarity between the first user and each classification reference user in category preference.

[0067] For example, in the potential customer mining scenario in car sales, the goal is to identify potential customers for three models of cars respectively. For each car, a user who has purchased the car can be selected as the classification benchmark user, or three virtual users corresponding to the three cars can be directly set as classification benchmark users as needed. According to the number of views of each user on different categories of content, the first similarity between each user and each classification benchmark user is determined, and then they are divided into reference groups corresponding to the classification benchmark users that match them. Among them, the classification benchmark user that matches a certain user can be a classification benchmark user whose first similarity with the user is within a preset threshold range. In this way, the users in a reference group are similar to the classification benchmark users of the group in their browsing category preferences. Therefore, the users in a reference group are more likely to buy the car corresponding to the classification benchmark user, and can be used as potential users of the car. Different users can be classified to facilitate differentiated marketing to users.

[0068] In addition, according to actual needs, a classification benchmark user can correspond to one or more reference groups. If the first similarity between a user and the classification benchmark user is within a certain threshold range, it is determined that the classification benchmark user matches the user. Among the multiple users corresponding to a classification benchmark user, the first similarity between each user and the user may be different. Therefore, if you want to further classify the users, you can divide the above-mentioned certain threshold range into multiple smaller ranges, each range corresponds to a reference group, and divide the multiple users matching a classification benchmark user into more refined reference groups.

[0069] For example, if the first similarity of two users is preset to 0-1, then the two users are considered similar, while 0-0.5 is generally similar, and 0.6-1 is extremely similar. If you only need to group users similar to the classification benchmark user into one reference group, that is, as long as they are similar, they are considered a match, then one classification benchmark user corresponds to one reference group, and the first similarity of each user in the reference group with the classification benchmark user is between 0-1; if you need to group extremely similar and generally similar users with the classification benchmark user into different reference groups, then two reference groups can be divided, and the first similarity of each user in one reference group with the classification benchmark user is between 0-0.5, and the first similarity of each user in the other reference group with the classification benchmark user is between 0.6-1, and the users who match the classification benchmark user are further divided into two more refined reference groups of extremely similar and generally similar.

[0070] S205, the electronic device determines the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user for each of the M categories, and obtains M second similarities corresponding to the first user and the matched classification reference user.

[0071] In a specific implementation, to determine the similarity between the first user and the matching classification benchmark user in each category, the data of the first user and the classification benchmark user can be mapped to points in the coordinate system, and the Euclidean distance between the first user and the classification benchmark user in each category can be calculated. The second similarity between the first user and the classification benchmark user is represented by the Euclidean distance, and the second similarity can be used to characterize the similarity in the degree of preference of the users for the content of the same category. For example, if the Euclidean distance is preset to 0-10 for very similar, 10-20 for relatively similar, and above 20 for dissimilar, the first user has browsed category A 10 times, and the classification benchmark user has browsed category A 15 times, according to the calculation formula of the Euclidean distance: Where p is the coordinate of the first user, q is the coordinate of the classification benchmark user, and when only one category of Euclidean distance is calculated, n = 1, and the calculated Euclidean distance is Thus, the second similarity between the first user and the classification benchmark user in category A is obtained, and the second similarity indicates that the first user and the classification benchmark user have similar preferences for category A. Since the value obtained by the Euclidean distance ranges from 0 to infinity, in actual applications, it is necessary to perform a secondary conversion between [0, 1] for convenience of calculation, using the following formula:

[0072] S206: The electronic device divides the first user into target groups corresponding to the M second similarities among the multiple target groups.

[0073] In S205 and S206, taking automobile marketing as an example, after users with similar browsing category preferences are divided into a group, users with similar category preferences can be further refined to determine the similarity between a user and the corresponding classification benchmark user in each category. For example, the first user, the second user and the classification benchmark user all browse more family and travel categories and less automobile technology categories in terms of browsing category preferences. When it comes to a specific category, the first user may browse the family category more frequently and the travel category less frequently, while the second user may browse the family category less frequently and the travel category more frequently. When classifying users, the first user and the second user can continue to be divided into different groups to further classify the users. With fine classification, in subsequent marketing, a car model that is more family-oriented can be recommended to the first user, and a car model that is more travel-oriented can be recommended to the second user, or different configurations of the car model corresponding to the classification benchmark user can be recommended to the first user and the second user respectively, and marketing activities can be arranged according to the user's preferences and needs. In addition, if the vehicle marketer wants to arrange marketing activities for a specific vehicle, the classification benchmark user can be a user who has already purchased the vehicle. Users who match the classification benchmark user are considered to have a higher possibility of purchasing the vehicle, and they can be the focus of potential customers for the vehicle during the marketing of the vehicle. Users who do not match the classification benchmark user are considered to have a relatively low possibility of purchasing the vehicle, and they can be temporarily excluded from consideration in the marketing activities of the vehicle.

[0074] It can be seen that in the embodiment of the present application, the electronic device first obtains the browsing behavior data of each user, and the browsing behavior data of the first user includes: the number of times the first user browses M categories, and then determines the first similarity between the first user and each classification benchmark user based on the number of times the first user browses the M categories, and further divides the first user into a reference group corresponding to the matching classification benchmark user based on the first similarity, and then determines the similarity between the first user and the matching classification benchmark user in each category based on the number of times the first user browses each of the M categories, and divides the first user into a target group corresponding to the M second similarities. It can be seen that according to the number of times the user browses the M categories, the first similarity and M second similarities between the user and the classification benchmark user are determined respectively, and the user is grouped twice based on the first similarity and the second similarity, which is conducive to accurate classification of users.

[0075] In a possible example, determining the first similarity between each user and each classification benchmark user based on the number of views of the first user for the M categories includes: establishing a multidimensional coordinate system; determining the first feature vector of the first user in the multidimensional coordinate system based on the number of views of the M categories; determining the second feature vector of each benchmark user in the multidimensional coordinate system; calculating the cosine similarity between the first feature vector and the second feature vector corresponding to each classification benchmark user, and using the cosine similarity as the first similarity between the first user and each classification benchmark user.

[0076] In a specific implementation, the dimensions of the established multidimensional coordinate system can correspond to M categories one by one, and the number of times a user browses a certain category can be reflected in the coordinate value of the dimension corresponding to the category in the coordinate system. For example, the number of times a user browses a certain category can be directly used as the coordinate value of the corresponding dimension, or when the number of browses is large, the number of browses N times corresponds to 1 unit on the coordinate. For example, there are three categories A, B, and C, and a three-dimensional coordinate system is established. User 1 browses category A 100 times, category B 50 times, and category C 10 times. The coordinate values ​​of the user corresponding to the three dimensions can be 100, 50, and 10, respectively, and the feature vector of user 1 can be expressed as (100, 50, 10), or the number of browses 10 times corresponds to one unit on the coordinate, and the feature vector of user 1 can be expressed as (10, 5, 1).

[0077] The formula for calculating the cosine similarity between the first eigenvector and a second eigenvector is: Among them, p, q are the first eigenvector and the second eigenvector respectively, and the value of i is equal to M. Cosine similarity measures the angle between two vectors. When the angle between the two vectors is 0 degrees, the cosine value is 1, when the angle is 90 degrees, the cosine value is 0, and when it is 180 degrees, the cosine value is -1. The design idea of ​​cosine similarity is that two vectors can be considered "similar" as long as they have the same direction, regardless of the degree of similarity.

[0078] Among them, when determining the classification benchmark user that matches the first user, when the cosine similarity between the first user and a certain classification benchmark user is within a certain threshold range, it is determined that the classification benchmark user matches the first user. For example, a classification benchmark user whose cosine similarity with the user is between 0-1 is set to match the user. The threshold range can be adjusted according to actual conditions. If a user has multiple classification benchmark users whose cosine similarities meet the conditions, the classification benchmark user with the highest similarity can be used as the matching benchmark user.

[0079] It can be seen that in this example, when determining the first similarity between the first user and each classified benchmark user, a multidimensional coordinate system is first established, and then the first feature vector of the first user in the multidimensional coordinate system and the second feature vector of each benchmark user in the multidimensional coordinate system are determined based on the number of views of M categories, and the cosine similarity between the first feature vector and each second feature vector is calculated. The data is mapped to the multidimensional coordinate system, and calculations are performed based on the corresponding vectors, which is conducive to improving the accuracy of determining the first similarity.

[0080] In one possible example, the dividing the first user into the target group corresponding to the M second similarities among the multiple target groups includes: determining a first classification standard for the first user based on the M second similarities corresponding to the first user; determining a first preset classification standard that matches the first classification standard from multiple preset classification standards, the multiple preset classification standards corresponding one-to-one to the multiple target groups; and dividing the first user into the target group corresponding to the first preset classification standard.

[0081] For example, with two categories A and B, the second similarity of 0-0.3 is divided into high similarity, and the second similarity of 0.4-1 is divided into low similarity. The multiple preset classification standards include: "A: high similarity, B: high similarity", corresponding to target group 1; "A: low similarity, B: high similarity", corresponding to target group 2; "A: high similarity, B: low similarity", corresponding to target group 3; "A: low similarity, B: low similarity", corresponding to target group 3. For example, if it is determined that the second similarities of the first user and the matching classification benchmark user in categories A and B are 0.2 and 0.5 respectively, the first classification standard of the first user can be "A: low similarity, B: high similarity". Therefore, the first preset classification standard corresponding to the first classification standard of the first user is "A: low similarity, B: high similarity", and the first user is classified into target group 2. The second similarity of each user in a target group and the matching classification benchmark user in each category meets the same classification standard.

[0082] It can be seen that in this example, the first classification standard of the first user is first determined based on the M second similarities corresponding to the first user, and then the first preset classification standard matching the first classification standard is determined from multiple preset classification standards, and then the first user is divided into a target group corresponding to the first preset classification standard. The second similarities of users in the same target group in each category meet the same classification standard, which is conducive to refined classification of users.

[0083] In one possible example, when the classification benchmark user is a user who has purchased a car, the method further includes: obtaining vehicle parameters of the car purchased by the classification benchmark user, the vehicle parameters including at least one of the following: model, color, size, performance, and price; obtaining a candidate car that matches the vehicle parameters of the car purchased by the classification benchmark user from a database, the database storing data on vehicles currently on sale by car sellers; and pushing the candidate car to the first user.

[0084] In a specific implementation, a candidate car may be the same car as a vehicle purchased by a classification benchmark user, or a car that is similar to the vehicle purchased by a classification benchmark user in various parameters. For example, if the purchased vehicle is a medium-sized family car, black in color, with good performance, and priced at RMB 250,000, then a medium-sized family car, dark in color, with good performance, and priced between RMB 200,000 and RMB 300,000 may be determined from the database as a candidate car and pushed to the first user.

[0085] In addition, since there are certain differences in the first classification standards corresponding to users in each target group corresponding to the same reference group, that is, there are differences in the second similarities between users in different target groups and classification benchmark users in each category, therefore, when obtaining candidate vehicles whose parameters are similar to vehicles purchased by classification benchmark users, the vehicle parameter range of the candidate vehicles can be determined according to the first classification standards corresponding to users in different target groups, that is, obtaining candidate vehicles that match the vehicle parameters of vehicles purchased by classification benchmark users from a database, including: obtaining the first classification standard corresponding to the first user, and obtaining candidate vehicles that match the vehicle parameters of vehicles purchased by classification benchmark users and the first classification standard from a database.

[0086] For example, if the vehicle purchased by the classification benchmark user is a medium-sized family car with a black color, if the first classification standard corresponding to the first user is obtained as: high similarity in the dark vehicle category and low similarity in the medium vehicle category, that is, the second similarity between the first user and the classification benchmark user in the dark vehicle category is high, and the second similarity in the medium vehicle category is low, when obtaining candidate vehicles, the vehicle model range can be expanded, while the vehicle color range can be relatively small, and the obtained candidate vehicles can be small-medium to medium-large family cars with a dark color. If the first classification standard corresponding to the first user is: low similarity in the dark vehicle category and high similarity in the medium vehicle category, the candidate vehicle can be a medium-sized family car with a dark or light color.

[0087] It can be seen that in this example, when the classification benchmark user is a user who has purchased a car, the vehicle parameters of the car purchased by the classification benchmark user can also be obtained, and then the candidate car matching the vehicle parameters of the car purchased by the classification benchmark user is obtained from the database and pushed to the first user. Since the first user and the classification benchmark user have similar browsing preferences, the candidate vehicle is more likely to meet the preferences and needs of the first user, which is conducive to detailed business activity arrangements based on the user classification results.

[0088] In one possible example, when the classification benchmark user is a user who has purchased a wealth management product, the method further includes: obtaining product parameters of the wealth management product purchased by the classification benchmark user, the product parameters including at least one of the following: product category, risk level, profitability, and capital liquidity; obtaining candidate wealth management products that match the product parameters of the wealth management product purchased by the classification benchmark user from a wealth management product database; and pushing the candidate wealth management products to the first user.

[0089] In a specific implementation, the candidate wealth management product may be a wealth management product that is the same as the wealth management product already purchased by the classification benchmark user, or a wealth management product that is similar to the wealth management product already purchased by the classification benchmark user in various parameters. For example, if the purchased wealth management product is a fund with a medium risk level, medium profitability, and high liquidity of funds, then a wealth management product whose product parameters meet the above standards can be identified in the database and pushed to the first user as a candidate wealth management product.

[0090] In addition, since there are certain differences in the first classification standards corresponding to users in each target group corresponding to the same reference group, that is, there are differences in the second similarities between users in different target groups and classification benchmark users in each category, therefore, when obtaining financial products whose parameters are similar to those of financial products purchased by classification benchmark users, the product parameter range of candidate financial products can be determined according to the first classification standards corresponding to users in different target groups, that is, candidate products that match the product parameters of financial products purchased by classification benchmark users are obtained from the database, including: obtaining the first classification standard corresponding to the first user, and obtaining candidate financial products that match the product parameters of financial products purchased by classification benchmark users and the first classification standard from the financial product database.

[0091] It can be seen that in this example, when the classification benchmark user is a user who has purchased financial products, it is also possible to obtain the product parameters of the financial products purchased by the classification benchmark user, and then obtain candidate financial products that match the financial products purchased by the classification benchmark user from the product database and push them to the first user. Since the first user and the classification benchmark user have similar browsing preferences, the candidate financial product is more likely to meet the preferences and needs of the first user, which is conducive to detailed business activity arrangements based on the user classification results.

[0092] In one possible example, when the classification benchmark user is a user who has purchased vehicle insurance, the method further includes: obtaining insurance information of the vehicle insurance purchased by the classification benchmark user, the insurance information including at least one of the following: insurance type, insurance premium, coverage amount, insurance company, and insurance company credit rating; obtaining candidate vehicle insurance that matches the insurance information of the vehicle insurance purchased by the classification benchmark user from a vehicle insurance database; and pushing the candidate vehicle insurance to the first user.

[0093] In a specific implementation, the candidate vehicle insurance may be the same vehicle insurance as the vehicle insurance purchased by the classification benchmark user, or the vehicle insurance whose insurance information is similar to the vehicle insurance purchased by the classification benchmark user. For example, the classification benchmark user purchased insurance product 1, the insurance type is full vehicle robbery and theft insurance, the insurance premium is 500 yuan per year, the guarantee amount is 10,000 yuan, the insurance company is Company A, and Company A has an excellent credit rating. In this case, the insurance product can be directly selected as the candidate vehicle insurance. Alternatively, a vehicle insurance with an annual insurance premium of 450-550 yuan, a guaranteed amount of 9,000-11,000 yuan, and an excellent credit rating of the insurance company in the full vehicle robbery and theft insurance can be determined and pushed to the first user as a candidate vehicle insurance.

[0094] In addition, since there are certain differences in the first classification criteria corresponding to users in each target group in multiple target groups corresponding to the same reference group, that is, there are differences in the second similarities between users in different target groups and classification benchmark users in each category, therefore, when obtaining candidate vehicle insurance with parameters similar to the vehicle insurance purchased by the classification benchmark user, the insurance information selection range of the candidate vehicle insurance can be determined according to the first classification criteria corresponding to users in different target groups, that is, obtaining candidate vehicles that match the insurance information of the vehicle insurance purchased by the classification benchmark user from the database, including: obtaining the first classification standard corresponding to the first user, and obtaining candidate vehicles that match the insurance information of the vehicle insurance purchased by the classification benchmark user and the first classification standard from the vehicle insurance database.

[0095] It can be seen that in this example, when the classification benchmark user is a user who has purchased vehicle insurance, the insurance information of the vehicle insurance purchased by the classification benchmark user can also be obtained, and then the candidate vehicle insurance that matches the insurance information of the vehicle insurance purchased by the classification benchmark user is obtained from the vehicle insurance database and pushed to the first user. Since the first user and the classification benchmark user have similar browsing preferences, the candidate vehicle insurance is more likely to meet the preferences and needs of the first user, which is conducive to refined business activity arrangements based on the user classification results.

[0096] In one possible example, when the classification benchmark user is a user who has purchased travel services, the method further includes: obtaining service information of the travel services purchased by the classification benchmark user, the service information including at least one of the following: travel destination, travel transportation method, travel duration, and travel cost composition; obtaining candidate travel services that match the service information of the travel services purchased by the classification benchmark user from a travel service database; and pushing the candidate travel services to the first user.

[0097] For example, the classification benchmark user's travel destination is A, the travel method is direct flight, the travel time is 3 days, and the travel expenses are 3,000 yuan, including 1,000 yuan for air tickets, 1,000 yuan for general hotel accommodation, and 1,000 yuan for food and beverage consumption. Then, the travel service with the same service information as the travel service purchased by the classification benchmark user can be taken as a candidate service and pushed to the first user. It can also be determined that the service information in the database includes: the travel destination is the same type of travel location as A, the travel method is direct public transportation, the travel duration is 2-7 days, the transportation cost in the travel cost is 900-1100 yuan, the accommodation is a general hotel with a daily cost of 200-400 yuan, and the food and beverage consumption is 900-1100 yuan. The travel service is determined as a candidate travel service and pushed to the first user.

[0098] In addition, since there are certain differences in the first classification criteria corresponding to users in each target group in multiple target groups corresponding to the same reference group, that is, there are differences in the second similarities between users in different target groups and classification benchmark users in each category, therefore, when obtaining travel services whose parameters are similar to the travel services purchased by classification benchmark users, the service information selection range of candidate travel services can be determined according to the first classification criteria corresponding to users in different target groups, that is, obtaining candidate travel services that match the service information of travel services purchased by classification benchmark users from the database, including: obtaining the first classification standard corresponding to the first user, and obtaining candidate vehicles that match the service information of travel services purchased by classification benchmark users and the first classification standard from the travel service database.

[0099] It can be seen that in this example, when the classification benchmark user is a user who has purchased travel services, it is also possible to obtain service information of the travel services purchased by the classification benchmark user, and then obtain candidate travel services that match the service information of the travel services purchased by the classification benchmark user from the travel service database and push them to the first user. Since the first user and the classification benchmark user have similar browsing preferences, the candidate travel service is more likely to meet the preferences and needs of the first user, which is conducive to refined business activity arrangements based on the user classification results.

[0100] The method of the embodiment of the present application is introduced above, and the device of the embodiment of the present application is introduced below.

[0101] See also Figure 3 , Figure 3 : is a schematic diagram of the structure of a user classification device provided in an embodiment of the present application, the device 30 includes:

[0102] The acquisition unit 301 is used to acquire browsing behavior data of each user of at least one user, where the browsing behavior data of a first user includes: browsing times of the first user for M categories, where M is a positive integer, and the first user is any one of the at least one user;

[0103] A first determining unit 302 is configured to determine a first similarity between the first user and each classification reference user according to the number of views of the first user on the M categories; and determine a classification reference user matching the first user according to the first similarity between the first user and each classification reference user;

[0104] A first grouping unit 303 is used to classify the first user into a reference group corresponding to the matched classification reference user, wherein the reference group corresponds to a plurality of target groups;

[0105] A second determining unit 304 is configured to determine the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user for each of the M categories, and obtain M second similarities corresponding to the first user and the matched classification reference user;

[0106] The second grouping unit 305 is configured to group the first user into a target group corresponding to the M second similarities among the multiple target groups.

[0107] In one possible example, in terms of determining the first similarity between the first user and each classification benchmark user based on the number of views of the first user for the M categories, the first determination unit 302 is specifically used to: establish a multidimensional coordinate system; determine the first feature vector of the first user in the multidimensional coordinate system based on the number of views of the M categories; determine the second feature vector of each classification benchmark user in the multidimensional coordinate system; calculate the cosine similarity between the first feature vector and the second feature vector corresponding to each classification benchmark user, and use the cosine similarity as the first similarity between the first user and each classification benchmark user.

[0108] In a possible example, the second grouping unit 305 is specifically used to: determine a first classification standard for the first user based on the M second similarities corresponding to the first user; determine a first preset classification standard that matches the first classification standard from multiple preset classification standards, and the multiple preset classification standards correspond one-to-one to the multiple target groups; and divide the first user into the target group corresponding to the first preset classification standard.

[0109] In a possible example, when the classification benchmark user is a user who has purchased a car, the device 30 also includes: a first push unit 306, used to obtain vehicle parameters of the vehicle purchased by the classification benchmark user, and the vehicle parameters include at least one of the following: model, color, size, performance, and price; obtain a candidate car that matches the vehicle parameters of the vehicle purchased by the classification benchmark user from a database, and the database stores data on vehicles currently on sale by car sellers; and push the candidate car to the first user.

[0110] In a possible example, when the classification benchmark user is a user who has purchased a financial product, the device 30 also includes: a second push unit 307, used to obtain product parameters of the financial product purchased by the classification benchmark user, and the product parameters include at least one of the following: product category, risk level, profitability, and capital liquidity; obtain candidate financial products that match the product parameters of the financial product purchased by the classification benchmark user from a financial product database; and push the candidate financial product to the first user.

[0111] In one possible example, when the classification benchmark user is a user who has purchased vehicle insurance, the device 30 also includes: a third push unit 308, used to obtain insurance information of the vehicle insurance purchased by the classification benchmark user, and the insurance information includes at least one of the following: insurance type, insurance premium, coverage amount, insurance company, and insurance company credit rating; obtaining candidate vehicle insurance that matches the insurance information of the vehicle insurance purchased by the classification benchmark user from a vehicle insurance database; and pushing the candidate vehicle insurance to the first user.

[0112] In one possible example, when the classification benchmark user is a user who has purchased travel services, the device 30 also includes: a fourth push unit 309, used to obtain service information of the travel service purchased by the classification benchmark user, and the service information includes at least one of the following: travel destination, travel transportation method, travel duration, and travel cost composition; obtaining candidate travel services that match the service information of the travel service purchased by the classification benchmark user from a travel service database; and pushing the candidate travel services to the first user.

[0113] It can be seen that in the embodiment of the present application, the user classification device can be implemented through each unit: first, the browsing behavior data of each user is obtained, and the browsing behavior data of the first user includes: the number of times the first user browses for M categories, and then the first similarity between the first user and each classification benchmark user is determined based on the number of times the first user browses for the M categories, and further based on the first similarity, the first user is divided into a reference group corresponding to the matching classification benchmark user, and then based on the number of times the first user browses for each of the M categories, the similarity between the first user and the matching classification benchmark user in each category is determined, and the first user is divided into a target group corresponding to the M second similarities. It can be seen that according to the number of times the user browses for the M categories, the first similarity and M second similarities between the user and the classification benchmark user are determined respectively, and the user is grouped twice according to the first similarity and the second similarity, which is conducive to accurate classification of users.

[0114] It should be noted that the above-mentioned units (acquisition unit 301, first determination unit 302, first grouping unit 303, second determination unit 304 and second grouping unit 305) are used to execute the relevant steps of the above-mentioned method. For example, acquisition unit 301 is used to execute the relevant content of S201, first determination unit 302 is used to execute the relevant content of S202 and S203, first grouping unit 303 is used to execute the relevant content of S204, second determination unit 304 is used to execute the relevant content of S205, and second grouping unit 305 is used to execute the relevant content of S206.

[0115] It can be understood that since the method embodiment and the device embodiment are different presentation forms of the same technical concept, the content of the method embodiment part in this application should be synchronously adapted to the device embodiment part and will not be repeated here.

[0116] In this embodiment, the user classification device 30 is presented in the form of a unit. The "unit" here may refer to an application-specific integrated circuit (ASIC), a processor and a memory that executes one or more software or firmware programs, an integrated logic circuit, and / or other devices that can provide the above functions.

[0117] See also Figure 4 , Figure 4 4 is a schematic diagram of the composition structure of an electronic device provided in an embodiment of the present application, wherein the electronic device 40 includes a processor 401, a memory 402, and a communication interface 403. The processor 401 is connected to the memory 402 and the communication interface 403, for example, the processor 401 can be connected to the memory 402 and the communication interface 403 via a bus.

[0118] The processor 401 may be a general purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the above program.

[0119] The communication interface 403 is used to communicate with other devices or communication networks, such as Ethernet, Radio Access Network (RAN), Wireless Local Area Networks (WLAN), etc.

[0120] The memory 402 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor via a bus. The memory may also be integrated with the processor.

[0121] The memory 402 is used to store application code for executing the above solution, and the execution is controlled by the processor 401. The processor 401 is used to execute the application code stored in the memory 402. The code stored in the memory 402 can execute any of the user classification methods provided above.

[0122] In one possible example, the code stored in the memory 402 is used to perform the following steps: obtaining browsing behavior data of each user in at least one user, the browsing behavior data of the first user including: the number of browsing times of the first user for M categories, where M is a positive integer, and the first user is any one of the at least one user; determining a first similarity between the first user and each classification benchmark user based on the number of browsing times of the first user for the M categories; determining a classification benchmark user matching the first user based on the first similarity corresponding to the first user and each classification benchmark user; dividing the first user into a reference group corresponding to the matching classification benchmark user, the reference group corresponding to multiple target groups; determining the similarity between the first user and the matching classification benchmark user in each category based on the number of browsing times of the first user for each category in the M categories, and obtaining M second similarities corresponding to the first user and the matching classification benchmark user; dividing the first user into a target group corresponding to the M second similarities among the multiple target groups.

[0123] In one possible example, in terms of determining the first similarity between the first user and each classification benchmark user based on the number of views of the first user for the M categories, the code stored in the memory 402 is specifically used to perform the following steps: establishing a multidimensional coordinate system; determining the first feature vector of the first user in the multidimensional coordinate system based on the number of views of the M categories; determining the second feature vector of each classification benchmark user in the multidimensional coordinate system; calculating the cosine similarity between the first feature vector and the second feature vector corresponding to each classification benchmark user, and using the cosine similarity as the first similarity between the first user and each classification benchmark user.

[0124] In one possible example, in terms of dividing the first user into the target group corresponding to the M second similarities among the multiple target groups, the code stored in the memory 402 is specifically used to perform the following steps: determine a first classification criterion for the first user based on the M second similarities corresponding to the first user; determine a first preset classification criterion that matches the first classification criterion from multiple preset classification criteria, the multiple preset classification criteria corresponding one-to-one to the multiple target groups; and divide the first user into the target group corresponding to the first preset classification criterion.

[0125] In one possible example, when the classification benchmark user is a user who has purchased a car, the code stored in the memory 402 is also used to perform the following steps: obtaining vehicle parameters of the vehicle purchased by the classification benchmark user, the vehicle parameters including at least one of the following: vehicle model, color, size, performance, and price; obtaining candidate vehicles that match the vehicle parameters of the vehicle purchased by the classification benchmark user from a database, the database storing data on vehicles currently on sale by car sellers; and pushing the candidate vehicles to the first user.

[0126] In one possible example, when the classification benchmark user is a user who has purchased a wealth management product, the code stored in the memory 402 is also used to perform the following steps: obtaining product parameters of the wealth management product purchased by the classification benchmark user, the product parameters including at least one of the following: product category, risk level, profitability, and liquidity of funds; obtaining candidate wealth management products that match the product parameters of the wealth management product purchased by the classification benchmark user from a wealth management product database; and pushing the candidate wealth management products to the first user.

[0127] In one possible example, when the classification benchmark user is a user who has purchased vehicle insurance, the code stored in the memory 402 is also used to perform the following steps: obtaining insurance information of the vehicle insurance purchased by the classification benchmark user, the insurance information including at least one of the following: insurance type, insurance premium, coverage amount, insurance company, and insurance company credit rating; obtaining candidate vehicle insurance that matches the insurance information of the vehicle insurance purchased by the classification benchmark user from a vehicle insurance database; and pushing the candidate vehicle insurance to the first user.

[0128] In one possible example, when the classification benchmark user is a user who has purchased travel services, the code stored in the memory 402 is also used to perform the following steps: obtaining service information of the travel service purchased by the classification benchmark user, the service information including at least one of the following: travel destination, travel transportation method, travel duration, and travel cost composition; obtaining candidate travel services that match the service information of the travel service purchased by the classification benchmark user from a travel service database; and pushing the candidate travel services to the first user.

[0129] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, and the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiments.

[0130] The present application also provides a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps of any method described in the above method embodiment. The computer program product may be a software installation package.

[0131] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0132] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0133] In the several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are only schematic, such as the division of the above-mentioned units, which is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical or other forms.

[0134] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0135] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0136] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a memory, including a number of instructions to enable a computer device (which can be a personal computer, server or network device, etc.) to perform all or part of the steps of the above-mentioned methods of each embodiment of the present application. The aforementioned memory includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, disk or CD-ROM and other media that can store program codes.

[0137] A person skilled in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable memory, and the memory can include: a flash drive, a read-only memory (English: Read-Only Memory, abbreviated as: ROM), a random access memory (English: Random Access Memory, abbreviated as: RAM), a magnetic disk or an optical disk, etc.

[0138] The embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core idea of ​​the present application. At the same time, for general technical personnel in this field, according to the idea of ​​the present application, there will be changes in the specific implementation method and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A user classification method, characterized in that: include: Obtain browsing behavior data of each user of at least one user, where the browsing behavior data of a first user includes: browsing times of the first user for M categories, where M is a positive integer, and the first user is any one of the at least one user; Determine, according to the browsing times of the first user and each classification reference user for the M categories, a first similarity between the first user and each classification reference user, wherein the first similarity is used to characterize the similarity between the first user and each classification reference user in category preference, wherein the classification reference user is a virtual user or an existing user; Determining a classification reference user matching the first user according to a first similarity between the first user and each classification reference user; Classifying the first user into a reference group corresponding to the matched classification reference user, wherein the reference group corresponds to a plurality of target groups; Determine the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user and each classification reference user for each category in the M categories, and obtain M second similarities corresponding to the first user and the matched classification reference user, wherein the second similarities are used to represent the similarity of the first user's preferences for content in the same category in the M categories; Determine a first classification standard for the first user based on the M second similarities corresponding to the first user; determine a first preset classification standard that matches the first classification standard from multiple preset classification standards, the multiple preset classification standards correspond one-to-one to the multiple target groups, and each preset classification standard includes M second similarity levels with the classification benchmark user in each category; and classify the first user into the target group corresponding to the first preset classification standard.

2. The method according to claim 1, characterized in that The determining, according to the browsing times of the first user for the M categories, a first similarity between the first user and each classification reference user comprises: Establish a multi-dimensional coordinate system; Determining a first feature vector of the first user in the multi-dimensional coordinate system according to the browsing times of the M categories; Determine a second feature vector of each classification reference user in the multi-dimensional coordinate system; The cosine similarity between the first feature vector and the second feature vector corresponding to each classification reference user is calculated, and the cosine similarity is used as the first similarity between the first user and each classification reference user.

3. The method according to claim 1 or 2, characterized in that: In the case where the classification reference user is a user who has purchased a car, the method further includes: Acquire vehicle parameters of the vehicle purchased by the classification reference user, wherein the vehicle parameters include at least one of the following: vehicle model, color, size, performance, and price; Acquire a candidate vehicle that matches vehicle parameters of the classification reference user's purchased vehicle from a database, wherein the database stores data on vehicles currently on sale by a car seller; The candidate vehicle is pushed to the first user.

4. The method according to claim 1 or 2, characterized in that: In the case where the classification basis user is a user who has purchased a financial product, the method further includes: Acquire product parameters of the wealth management products purchased by the classification basis user, wherein the product parameters include at least one of the following: product category, risk level, profitability, and liquidity of funds; Acquire, from a wealth management product database, candidate wealth management products that match product parameters of wealth management products purchased by the classification basis user; The candidate financial product is pushed to the first user.

5. The method according to claim 1 or 2, characterized in that: In the case where the classification reference user is a user who has purchased vehicle insurance, the method further includes: Obtaining insurance information of vehicle insurance purchased by the classification base user, wherein the insurance information includes at least one of the following: insurance type, insurance cost, coverage amount, insurance company, and insurance company credit rating; Acquire, from a vehicle insurance database, candidate vehicle insurance that matches insurance information of vehicle insurance purchased by the classification reference user; The candidate vehicle insurance is pushed to the first user.

6. The method according to claim 1 or 2, characterized in that: In the case where the classification reference user is a user who has purchased travel services, the method further includes: Acquire service information of the travel service purchased by the classification basis user, wherein the service information includes at least one of the following: travel destination, travel mode, travel duration, and travel cost composition; Acquire, from a travel service database, candidate travel services that match the service information of travel services purchased by the classification reference user; The candidate travel service is pushed to the first user.

7. A user classification device, characterized in that: include: an acquisition unit, configured to acquire browsing behavior data of each user of at least one user, wherein the browsing behavior data of a first user includes: browsing times of the first user for M categories, wherein M is a positive integer, and the first user is any one of the at least one user; A first determining unit is configured to determine a first similarity between the first user and each classification reference user according to the number of views of the first user and each classification reference user for the M categories, wherein the first similarity is used to characterize the similarity between the first user and each classification reference user in category preference; and to determine a classification reference user that matches the first user according to the first similarity corresponding to the first user and each classification reference user, wherein the classification reference user is a virtual user or an existing user; A first grouping unit, configured to classify the first user into a reference group corresponding to the matched classification reference user, wherein the reference group corresponds to a plurality of target groups; a second determining unit, configured to determine the similarity between the first user and the matched classification reference user in each category according to the number of views of the first user and each classification reference user for each category in the M categories, and obtain M second similarities corresponding to the first user and the matched classification reference user, wherein the second similarities are used to represent the similarity of the first user's preferences for content in the same category in the M categories; A second grouping unit is used to determine a first classification standard for the first user based on the M second similarities corresponding to the first user; determine a first preset classification standard that matches the first classification standard from multiple preset classification standards, the multiple preset classification standards correspond one-to-one to the multiple target groups, each target group includes M second similarity levels with the classification benchmark user in each category; and classify the first user into the target group corresponding to the first preset classification standard.

8. An electronic device, characterized in that: The method comprises a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the programs include instructions for executing the steps in the method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: A computer program for electronic data exchange is stored, wherein the computer program enables a computer to execute the method according to any one of claims 1 to 6.