An information recommendation method, device, equipment and readable storage medium

Abstract

This application discloses an information recommendation method, apparatus, device, and readable storage medium. A tag classification library is pre-created, containing several tag types and user information belonging to each tag type. A user's tag type is determined based on their historical profile. Then, based on the acquired current operational strategy, several target sets are created, each with several tag types. For each target set, users matching several tag types are selected from the tag classification library as target users according to the set's tag types. The target user information is stored in the target set. Information matching the tag types set for each target set is sent to the target users in each target set, thereby achieving the push of specific messages to specific users.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to an information recommendation method, apparatus, device, and readable storage medium. Background Technology

[0002] With the continuous development of internet technology, merchants are focusing on developing online platforms to meet current user needs. When users use these platforms, their information is recorded based on their login accounts. Recommendation messages can be sent periodically to users who have already used the platform. However, due to the current chaotic user information management, recommendation messages are not differentiated from individual users. This can lead to users receiving many irrelevant messages, resulting in user churn. Therefore, how to effectively recommend content is a persistent concern. Summary of the Invention

[0003] In view of this, this application provides an information recommendation method, apparatus, device, and readable storage medium to facilitate information recommendation.

[0004] To achieve the above objectives, the following solution is proposed:

[0005] An information recommendation method, comprising:

[0006] Obtain the current operational strategy;

[0007] Based on the aforementioned operational strategy, several target sets are created, and each target set is assigned several tag types;

[0008] Obtain a pre-created tag classification library, which contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles.

[0009] For each target set, according to the several tag types set for the target set, users who match the several tag types are selected from the tag classification library as target users, and the information of the target users is stored in the target set;

[0010] Send information that matches the tag type set for each target set to the target users in each target set.

[0011] Optionally, the process of creating the tag classification library includes:

[0012] Based on each user's historical profile, determine the corresponding tag type for each user;

[0013] A tag classification library is created using the tag type corresponding to each user. The tag classification library contains several tag types and user information belonging to each tag type.

[0014] Optionally, before determining the tag type for each user based on their historical profile, the following steps are also included:

[0015] Obtain business requirements;

[0016] The process of determining the tag type for each user based on their historical profile includes:

[0017] Based on the aforementioned business requirements, the historical profile of each user is filtered to obtain the filtered historical profile of each user.

[0018] By utilizing the historical profiles of each filtered user, the corresponding tag type for each user is determined.

[0019] Optionally, after creating the tag classification library, the following may also be included:

[0020] Update each user's historical profile according to a pre-set time period;

[0021] Based on the updated historical profile of each user, determine the new tag type for each user;

[0022] The tag classification library is updated using the new tag type corresponding to each user.

[0023] Optional, also includes:

[0024] Get the updated tag classification library;

[0025] For each target set, according to the several tag types set for the target set, users who match the several tag types are selected from the updated tag classification library as new target users, and the information of the original target users in the target set is replaced with the information of the new target users.

[0026] An information recommendation device, comprising:

[0027] The operation strategy acquisition module is used to acquire the current operation strategy;

[0028] The target set creation module is used to create several target sets according to the operational strategy, and each target set is configured with several tag types;

[0029] The category library acquisition module is used to acquire a pre-created tag category library, which contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles.

[0030] The target user filtering module is used to filter users who match the several tag types set in the tag classification library for each target set as target users, and store the information of the target users in the target set.

[0031] The information sending module is used to send information that matches the tag type set for each target set to the target users in each target set.

[0032] Optional, also includes:

[0033] The category library creation module is used to create the tag category library, wherein the process of creating the tag category library includes:

[0034] Based on each user's historical profile, determine their corresponding tag type;

[0035] A tag classification library is created using the tag type corresponding to each user. The tag classification library contains several tag types and user information belonging to each tag type.

[0036] Optionally, the classification library creation module is also used to obtain business requirements before executing the process of determining the tag type corresponding to each user based on each user's historical profile;

[0037] The classification library creation module performs a process based on each user's historical profile to determine the corresponding tag type for each user, including:

[0038] Based on the aforementioned business requirements, the historical profile of each user is filtered to obtain the filtered historical profile of each user.

[0039] By utilizing the historical profiles of each filtered user, the corresponding tag type for each user is determined.

[0040] An information recommendation device includes: a memory and a processor;

[0041] The memory is used to store programs;

[0042] The processor is used to execute the program to implement the various steps of the information recommendation method described above.

[0043] A readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the information recommendation method described above.

[0044] As can be seen from the above technical solutions, the information recommendation method, apparatus, device, and readable storage medium provided in this application pre-create a tag classification library, wherein the tag classification library contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles. Then, according to the obtained current operation strategy, several target sets are created, wherein each target set is set with several tag types. For each target set, according to the several tag types set in the target set, users who match several tag types in the tag classification library are selected as target users, and the information of the target users is stored in the target set. Information matching the tag types set in each target set is sent to the target users in each target set, thereby realizing the push of specific messages to specific users.

[0045] Furthermore, in this application, by pre-creating a tag classification library, users are first classified according to tag type. After creating a target set according to the operation strategy, target users that match the tag type can be filtered based on the tag set in the tag classification library according to the tag type set set in the target set. Compared with creating a target set according to the operation strategy and then judging whether each user matches the tag type set set in the target set, the amount of data processed is smaller, which can improve the efficiency of information recommendation to a certain extent. Attached Figure Description

[0046] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0047] Figure 1 A flowchart of an information recommendation method provided in an embodiment of this application;

[0048] Figure 2 This is a schematic diagram of an information recommendation device provided in an embodiment of this application;

[0049] Figure 3 This is a hardware structure block diagram of an information recommendation device disclosed in an embodiment of this application. Detailed Implementation

[0050] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0051] Figure 1 A flowchart of an information recommendation method provided in this application embodiment is shown below. Figure 1 As shown, the method may include the following steps:

[0052] Step S100: Obtain the current operational strategy.

[0053] Specifically, an operational strategy can be the means or methods adopted to achieve certain goals within a specific period. Different products generally have different lifecycles, different competitors, and different resources, all of which influence operational strategies.

[0054] In actual operation, an operational strategy aimed at driving first-time purchases will target active users who have never bought anything before; an operational strategy aimed at driving repeat purchases will target active users who have bought things before; and an operational strategy aimed at reactivating consumption will target users who have bought things before but have not bought anything recently.

[0055] Step S101: Create several target sets according to the operation strategy.

[0056] Each of the above target sets has several label types.

[0057] Specifically, based on the operational strategies obtained in the above steps, several target sets can be created. For example, assuming the operational strategy is to encourage VIP customers to make repeat purchases, several target sets can be created. One target set can be tagged with "VIP users" or "users interested in discount or full-reduction products," while another target set can be tagged with "VIP users" or "users interested in new product launches." Other target sets can also be created.

[0058] Step S102: Obtain the pre-created tag classification library.

[0059] The aforementioned tag classification library may contain several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles.

[0060] Specifically, the pre-created tag classification library can contain several tag types, such as "VIP User," "User Following Discounted or Full-Price Products," and "User Following New Product Launches." Each tag type can be associated with user information belonging to that tag type. For example, based on user A's historical profile, it can be determined that user A is a VIP user and follows new product launches. In this case, user A's information can be placed in the tag classification library and associated with the "VIP User" and "User Following New Product Launches" tag types, but not with the "User Following Discounted or Full-Price Products" tag type. Thus, when selecting the "VIP User" tag type from the tag classification library, user A's information can be obtained.

[0061] Step S103: For each target set, according to the several tag types set for the target set, select users who match the several tag types in the tag classification library as target users, and store the information of the target users into the target set.

[0062] Specifically, the users in the target set obtained through these steps are all users who match the tag types defined in the target set. Moreover, by filtering users who match the tag types from the tag classification library, compared to comparing each user's personal tag with the defined tag types one by one to determine the target users, the amount of data processed is smaller, thereby improving the efficiency of information recommendation to some extent.

[0063] For example, if the target set is defined with tag types "tag type 1", "tag type 2", and "tag type 3", then according to the target set's "tag type 1", "tag type 2", and "tag type 3", find "tag type 1", "tag type 2", and "tag type 3" in the tag classification library. Also find the information of user A, user B, and user C associated with "tag type 1", the information of user A and user C associated with "tag type 2", and the information of user A associated with "tag type 3". Then, take the intersection of the information of the users associated with the above three tag types to determine the target user as user A. Finally, store user A's information in the target set. The determined user A is a user who matches the above three tag types.

[0064] Step S104: Send information that matches the tag type set for each target set to the target users in each target set.

[0065] Specifically, in the above steps, after determining the information of the users included in the target set, information matching the tag type set for each target set can be sent to the target users in each target set based on the information of the target users.

[0066] For example, based on the operational strategy, it is necessary to encourage VIP users to make repeat purchases. Therefore, a target set is created with the tag types set as "VIP users" and "users who are interested in new product launches". Since the information of users whose tag types match the target set has been put into the target set, information that matches the new product launch can be sent to the target users, thereby achieving precise operation.

[0067] In the above embodiments, an information recommendation method is provided. This method involves pre-creating a tag classification library, which contains several tag types and user information belonging to each tag type. The tag type a user belongs to is determined based on their historical profile. Then, based on the acquired current operational strategy, several target sets are created, each with several tag types. For each target set, users matching several tag types are selected from the tag classification library as target users according to the set's tag types. The target user information is stored in the target set. Information matching the tag types set in each target set is sent to the target users in each target set, thereby enabling the push of specific messages to specific users.

[0068] Furthermore, in this application, by pre-creating a tag classification library, users are first classified according to tag type. After creating a target set according to the operation strategy, target users that match the tag type can be filtered based on the tag set in the tag classification library according to the tag type set set in the target set. Compared with creating a target set according to the operation strategy and then judging whether each user matches the tag type set set in the target set, the amount of data processed is smaller, which can improve the efficiency of information recommendation to a certain extent.

[0069] In some embodiments of this application, the process of creating a tag classification library is described, which may include:

[0070] S11. Based on each user's historical profile, determine the corresponding tag type for each user.

[0071] Specifically, the first step is to obtain the historical profile of each user. Based on the historical profile of each user, the corresponding tag type for each user can be determined. There can be many tag types for each user. For example, in terms of geography, the tag type can be divided into "Guangdong", "Guangxi" and "Fujian", etc. In terms of coupons, the tag type can be divided into "holding coupons" and "not holding coupons", etc. In terms of discount sensitivity, the tag type can be divided into "high", "medium" and "low", etc.

[0072] S12. Create a tag classification library using the tag type corresponding to each user.

[0073] The aforementioned tag classification library contains several tag types, as well as user information belonging to each tag type.

[0074] Specifically, the above steps can determine the tag type corresponding to each user. After determining the tag type of each user, the tag types corresponding to all users are counted and deduplicated, and finally all tag types can be obtained. Then, for each obtained tag type, users with that tag type are selected, and the tag type is associated with the user, thereby completing the creation of the tag classification library.

[0075] For example, suppose that user A, whose tag type is determined through the aforementioned steps, is "Guangdong", "Male", and "Medium Discount Sensitivity"; user B's tag type is "Guangdong", "Male", and "High Discount Sensitivity"; and user C's tag type is "Fujian", "Male", and "Medium Discount Sensitivity". This yields five tag types: "Guangdong", "Fujian", "Male", "Medium Discount Sensitivity", and "High Discount Sensitivity". Then, for the tag type "Guangdong", find users with this tag type, namely user A and user B. In the tag classification library, associate the tag type "Guangdong" with user A and user B. Therefore, by filtering for the tag type "Guangdong" in the tag classification library, you can obtain the information of user A and user B associated with "Guangdong".

[0076] Since the historical profile of each user contains a large amount of data, much of the information may be unnecessary later. If all the data is obtained, a large amount of invalid data may be generated. Therefore, in some embodiments of this application, business requirements can be obtained before executing step S11, which determines the tag type corresponding to each user based on the historical profile of each user.

[0077] Specifically, after obtaining business requirements, determining the tag type for each user based on their historical profile can include: filtering each user's historical profile according to business requirements to obtain a filtered historical profile for each user, and using the filtered historical profile for each user to determine the tag type for each user.

[0078] The business requirements are determined based on the current plan. Based on the determined business requirements, the historical profiles of each user can be initially filtered to retain only the data that is helpful to the current business requirements, thereby reducing data storage.

[0079] For example, suppose the current business need is to boost the sales of product A. However, the use of product A is the same regardless of the province or the user's gender. Therefore, we can filter each user's historical profile, removing data related to region and gender, thus reducing the amount of data to be processed and improving processing efficiency.

[0080] Because user profiles change in real time during actual operation, the tag type for each user changes, making the information in the created tag classification library inaccurate and hindering accurate user identification. Therefore, in some embodiments of this application, after executing step S12 and creating a tag classification library using the tag type for each user, the following may be included:

[0081] S21. Update the historical profile of each user according to the preset time period.

[0082] Specifically, since the historical profile of each user changes little in a short period of time, the impact on each user's tag type is small. Therefore, a time period can be set to update the historical profile of each user based on the actual operation situation.

[0083] S22. Based on the updated historical profile of each user, determine the new tag type for each user.

[0084] Specifically, by using the updated historical profile of each user through the above steps, a new tag type can be determined for each user. Alternatively, a new tag type can be generated for each user based on the updated historical profile, directly overwriting the existing tag type. Another approach is to first determine the updated data in the historical profile, then obtain the corresponding new tag type based on the updated data, and replace the original tag type with the new tag type.

[0085] S23. Update the tag classification library using the new tag type corresponding to each user.

[0086] Specifically, after obtaining the new tag type for each user through the above steps, the tag classification library can be updated using the new tag type for each user. During the update process, first, for each user, determine the tag type before and after the change. Then, in the tag classification library, first find the user's tag type before the change, delete the associated information for that user, and then find the user's tag type after the change, associating it with that user's information. If, after deleting the user's information, there is no other user information associated with the tag type before the change, then that tag type can be deleted from the tag classification library.

[0087] In the above embodiments, since the updated tag classification library is obtained based on the updated user historical profile, the tag type to which the user currently belongs can be accurately determined.

[0088] Furthermore, to ensure that all users in the target set are target users who conform to the several tag types set in the target set, and to guarantee the accuracy of information recommendation, in some embodiments of this application, after the tag classification library is updated, the following may also be included:

[0089] S31. Obtain the updated tag classification library.

[0090] Specifically, the tag classification library can be updated according to the update steps provided in the above embodiments.

[0091] S32. For each target set, according to the several tag types set in the target set, select users who match the several tag types in the updated tag classification library as new target users, and replace the information of the original target users in the target set with the information of the new target users.

[0092] Specifically, using the updated tag classification library obtained in the above steps, new target users can be selected from the updated tag classification library for each target set, and the information of the new target users can replace the original target user information in the target set. The user information in the updated target set consists of user information that conforms to several tag types defined for the target set.

[0093] In the above embodiments, by obtaining the updated tag classification library, each target set is updated so that the users in each target set are target users who meet the several tag types set for that target set, thereby ensuring the accuracy of information recommendation to a certain extent.

[0094] The following describes an information recommendation device provided by an embodiment of this application. The information recommendation device described below can be referred to in correspondence with the information recommendation method described above.

[0095] Figure 2 This application provides a schematic diagram of an information recommendation device, which may include:

[0096] Operational strategy acquisition module 10 is used to acquire the current operational strategy;

[0097] The target set creation module 20 is used to create several target sets according to the operation strategy, and each target set is configured with several tag types;

[0098] The category library acquisition module 30 is used to acquire a pre-created tag category library, which contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles.

[0099] The target user filtering module 40 is used to filter users who match the several tag types set in the tag classification library for each target set as target users, and store the information of the target users in the target set.

[0100] The information sending module 50 is used to send information that matches the tag type set for each target set to the target users in each target set.

[0101] In this application, by pre-creating a tag classification library, users are first classified according to tag type. After the target set creation module 20 creates the target set according to the operation strategy, the target user filtering module 40 can filter users who match the tag types set in the tag classification library as target users according to the several tag types set in the target set, and store the information of the target users in the target set. Compared with creating the target set according to the operation strategy and then judging whether each user matches the tag type set in the target set, the amount of data processed is smaller, which can improve the efficiency of information recommendation to a certain extent.

[0102] Optionally, the information recommendation device may also include:

[0103] The category library creation module is used to create the tag category library, wherein the process of creating the tag category library may include:

[0104] Based on each user's historical profile, determine their corresponding tag type;

[0105] A tag classification library is created using the tag type corresponding to each user. The tag classification library contains several tag types and user information belonging to each tag type.

[0106] Optionally, the classification library creation module is also used to obtain business requirements before executing the process of determining the tag type corresponding to each user based on each user's historical profile;

[0107] The classification library creation module performs a process of determining the tag type corresponding to each user based on each user's historical profile, which may include:

[0108] Based on the aforementioned business requirements, the historical profile of each user is filtered to obtain the filtered historical profile of each user.

[0109] By utilizing the historical profiles of each filtered user, the corresponding tag type for each user is determined.

[0110] Optionally, the information recommendation device may also include:

[0111] The historical profile update module is used to update the historical profile of each user according to a preset time period.

[0112] The tag type update module is used to determine the new tag type for each user based on the updated historical profile of each user;

[0113] The tag classification library update module is used to update the tag classification library using the new tag types corresponding to each user.

[0114] Optionally, the information recommendation device may also include:

[0115] The new category library acquisition module is used to acquire the updated tag category library;

[0116] The target set update module is used to, for each target set, select users who match the several tag types set in the updated tag classification library as new target users according to the several tag types set in the target set, and replace the information of the original target users in the target set with the information of the new target users.

[0117] This application also provides an information recommendation device. Figure 3 The hardware structure block diagram of the information recommendation device is shown below. Figure 3 The hardware structure of the information recommendation device may include: at least one processor 1, at least one communication interface 2, at least one memory 3, and at least one communication bus 4;

[0118] In this embodiment, the number of processor 1, communication interface 2, memory 3, and communication bus 4 is at least one, and processor 1, communication interface 2, and memory 3 communicate with each other through communication bus 4.

[0119] Processor 1 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention.

[0120] Memory 3 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk storage device;

[0121] The memory stores a program, which the processor can call. The program is used to implement the various processing steps in the aforementioned information recommendation method.

[0122] This application embodiment also provides a storage medium that can store a program suitable for execution by a processor, the program being used to implement the various processing flows in the aforementioned information recommendation method.

[0123] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0124] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined with each other, and the same or similar parts can be referred to each other.

[0125] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An information recommendation method characterized by comprising: include: Obtain the current operational strategy; Based on the aforementioned operational strategy, several target sets are created, and each target set is assigned several tag types; Obtain a pre-created tag classification library, which contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles. For each target set, according to the several tag types set for the target set, users who match the several tag types are selected from the tag classification library as target users, and the information of the target users is stored in the target set; Send information that matches the tag type set for each target set to the target users in each target set; The process of creating the tag classification library includes: Obtain business requirements; Based on the aforementioned business requirements, the historical profile of each user is filtered to obtain the filtered historical profile of each user. By utilizing the historical profiles of each filtered user, the corresponding tag type for each user can be determined; A tag classification library is created using the tag type corresponding to each user. The tag classification library contains several tag types and user information belonging to each tag type. After creating the tag classification library, the following is also included: Update each user's historical profile according to a pre-set time period; Based on the updated historical profile of each user, determine the new tag type for each user; Update the tag classification library using the new tag type corresponding to each user; Also includes: Get the updated tag classification library; For each target set, according to the several tag types set for the target set, users who match the several tag types are selected from the updated tag classification library as new target users, and the information of the original target users in the target set is replaced with the information of the new target users.

2. An information recommendation device characterized by comprising: include: The operation strategy acquisition module is used to acquire the current operation strategy; The target set creation module is used to create several target sets according to the operational strategy, and each target set is configured with several tag types; The category library acquisition module is used to acquire a pre-created tag category library, which contains several tag types and user information belonging to each tag type. The tag type to which a user belongs is determined based on their respective historical profiles. The target user filtering module is used to filter users who match the several tag types set in the tag classification library for each target set as target users, and store the information of the target users in the target set. The information sending module is used to send information that matches the tag type set for each target set to the target users in each target set; The category library creation module is used to create the tag category library, and the process of creating the tag category library includes: Obtain business requirements; Based on the aforementioned business requirements, the historical profile of each user is filtered to obtain the filtered historical profile of each user. By utilizing the historical profiles of each filtered user, the corresponding tag type for each user can be determined; A tag classification library is created using the tag type corresponding to each user. The tag classification library contains several tag types and user information belonging to each tag type. The information recommendation device further includes: The historical profile update module is used to update the historical profile of each user according to a preset time period. The tag type update module is used to determine the new tag type for each user based on the updated historical profile of each user; The tag classification library update module is used to update the tag classification library using the new tag types corresponding to each user; The new category library acquisition module is used to acquire the updated tag category library; The target set update module is used to, for each target set, select users who match the several tag types set in the updated tag classification library as new target users according to the several tag types set in the target set, and replace the information of the original target users in the target set with the information of the new target users.

3. An information recommendation device characterized by comprising: include: Memory and processor; The memory is used to store programs; The processor is used to execute the program to implement the various steps of the information recommendation method as claimed in claim 1.

4. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the various steps of the information recommendation method as claimed in claim 1.

Images