Information push methods, devices, equipment and storage media

Abstract

This application provides an information push method, apparatus, device, and storage medium. The method includes: acquiring pet information of a target entity pet and object information of a target object corresponding to the target entity pet; searching for an object cluster corresponding to the target object based on the pet information of the target entity pet and the object information of the target object; wherein the object cluster includes a first object corresponding to a first pet in the physical pet cluster corresponding to the target entity pet, a second object corresponding to a second pet in the virtual pet cluster corresponding to the target entity pet, and the target object; acquiring a set of outfit data for the object cluster; acquiring an outfit push message for the target entity pet based on the outfit data set; and sending the outfit push message to the target object. Using this application embodiment can improve the accuracy of pet outfit push and is beneficial for improving the matching effect.

Description

Technical Field

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

[0002] As material needs are met, pet ownership has become a way to fulfill emotional needs, and users enjoy dressing up their pets in clothes and accessories. Currently, offline styling requires users to select sets or individual items one by one until they are satisfied. However, pets may not cooperate during the styling process. Online selections of clothing or accessories may not produce satisfactory results, leading to high return rates. Summary of the Invention

[0003] This application provides an information push method, apparatus, device, and storage medium, which can improve the accuracy of pet outfit recommendations and improve the matching effect.

[0004] This application provides an information push method, comprising: acquiring pet information of a target physical pet and object information of a target object corresponding to the target physical pet; based on the pet information of the target physical pet and the object information of the target object, searching for an object cluster corresponding to the target object; the object cluster includes a first object corresponding to a first pet in the physical pet cluster corresponding to the target physical pet, a second object corresponding to a second pet in the virtual pet cluster corresponding to the target physical pet, and the target object; acquiring a set of outfit data for the object cluster; acquiring an outfit push message for the target physical pet based on the outfit data set; and sending the outfit push message to the target object. Thus, the outfit push information for the target physical pet references the target object's previous outfit data, as well as the outfit data of other objects feeding physical pets and virtual pets corresponding to the target object, which can improve the accuracy of pet outfit push and is beneficial to improving the matching effect.

[0005] In a possible example, based on a set of outfit data, obtaining the outfit push message for the target entity's pet includes: obtaining the pet outfit fusion feature of the target object based on the outfit data set of the object cluster; obtaining the matching value between the pet outfit feature and the pet outfit fusion feature of each pet outfit in the set of pet outfits to be pushed; selecting the target pet outfit for the target entity's pet from the pet outfit set based on the magnitude of the matching value; and obtaining the outfit push message for the target pet outfit. In other words, selecting the target pet outfit for the target entity's pet from the set of pet outfits based on the magnitude of the matching value between the outfit feature and the pet outfit fusion feature of each pet outfit in the set of pet outfits to be pushed can improve the accuracy of obtaining the outfit push message for the target pet, thus improving the push effect and the matching effect.

[0006] In one possible example, based on a clothing data set, the pet clothing fusion feature of the target object is obtained, including: obtaining the pet clothing preference features of the target object, the first object, and the second object respectively based on the clothing data set; obtaining the object weights of the target object, the first object, and the second object respectively; and obtaining the pet clothing fusion feature of the target object based on the pet clothing preference features and object weights of the target object, the first object, and the second object. Thus, by obtaining the pet clothing fusion feature of the object cluster based on the object weights and pet clothing preference features of each object in the object cluster, the accuracy of obtaining the pet clothing fusion feature can be improved, which is beneficial to improving the accuracy of selecting the target pet's clothing.

[0007] In one possible example, the outfit dataset includes pet outfit data for each of multiple first pet outfits for the target object. Based on the outfit dataset, the pet outfit preference features of the target object, the first object, and the second object are obtained, including: obtaining the target object's preference value for the first pet outfits; selecting at least two second pet outfits from the multiple first pet outfits based on the magnitude of the preference value; obtaining pet outfit preference sub-features for the second pet outfits based on their pet outfit data; and obtaining the target object's pet outfit preference features based on the outfit weights and pet outfit preference sub-features of the second pet outfits. In other words, analyzing the target object's pet outfit preference features from the perspective of the target object's preferred pet outfit sub-features and outfit weights can improve the accuracy of obtaining pet outfit preference features, which is beneficial for improving the accuracy of pushing target pet outfits.

[0008] In one possible example, the pet outfit data for the second pet outfit includes scenario data of the target audience purchasing the second pet outfit. The push notification method further includes: determining the target audience's pet outfit purchase scenario based on the scenario data; and sending an outfit push notification message to the target audience when the pet outfit purchase scenario occurs. This can further improve the accuracy of the delivery data and enhance the effectiveness of the campaign.

[0009] In a possible example, the object weights of the target object, the first object, and the second object are obtained separately, including: obtaining a first quantity of the first object's clothing data subset, a second quantity of the second object's clothing data subset, and a third quantity of the target object's clothing data subset; based on the third quantity, the first quantity, and the second quantity, the object weight of the target object is obtained; a first similarity value is obtained between the first object and the target object, and a second similarity value is obtained between the second object and the target object; based on the first and second similarity values, and the object weight of the target object, the object weights of the first object and the second object are obtained separately. In other words, determining the object weight of the target object first, and then obtaining the object weights of the first and second objects based on the similarity values ​​between the first and second objects and the target object, can improve the accuracy of weight allocation and is beneficial for improving the accuracy of obtaining the pet clothing fusion features of the target object.

[0010] In one possible example, the target pet's outfit includes a first outfit accessory and a second outfit accessory. The information push method further includes: receiving a replacement request for the first outfit accessory from the target object; searching for a third outfit accessory corresponding to the first and second outfit accessories from a pre-stored outfit accessory library based on pet outfit fusion features; obtaining a reference outfit image of the target pet based on the third and second outfit accessories; and sending the reference outfit image to the target object. In this way, the target object can determine whether to purchase the outfit accessory by comparing the outfit effect images of the replaced and unreplaced outfit accessories, thus improving push efficiency.

[0011] One embodiment of this application provides an information push device, including:

[0012] The processing unit is used to obtain the pet information of the target entity pet and the object information of the target object corresponding to the target entity pet; based on the pet information of the target entity pet and the object information of the target object, to find the object cluster corresponding to the target object; the object cluster includes the first object corresponding to the first pet in the entity pet cluster corresponding to the target entity pet, the second object corresponding to the second pet in the virtual pet cluster corresponding to the target entity pet, and the target object; to obtain the outfit data set of the object cluster; and based on the outfit data set, to obtain the outfit push message of the target entity pet.

[0013] The communication unit is used to send outfit recommendation messages to the target object.

[0014] One aspect of this application provides a computer device, including a memory and a processor connected to the memory. The memory stores a computer program, and the processor invokes the computer program to cause the computer device to execute the method provided in one aspect of this application.

[0015] One aspect of this application provides a computer-readable storage medium. This computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor, causing a computer device having a processor to perform the method provided in one aspect of this application.

[0016] According to one aspect of this application, a computer program product is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method provided in the above aspect.

[0017] This application embodiment can first locate the object cluster corresponding to the target object based on the pet information of the target physical pet and the object information of the target object corresponding to the target physical pet. Then, based on the outfit data set of the object cluster, it obtains the outfit push message for the target physical pet. The object cluster includes the target object, the first object corresponding to the first pet in the physical pet cluster corresponding to the target physical pet, and the second object corresponding to the second pet in the virtual pet cluster corresponding to the target physical pet. Thus, the outfit push information for the target physical pet references the target object's previous outfit data, as well as the outfit data of other objects feeding physical pets and virtual pets corresponding to the target object, which can improve the accuracy of pet outfit pushes. After sending the outfit push message to the target object, the target object can purchase the outfit accessories in the outfit push message, which helps improve the overall outfit effect. Attached Figure Description

[0018] 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of a network architecture provided in an embodiment of this application;

[0020] Figure 2 This application provides an illustration of a scenario for purchasing clothing accessories for a physical pet.

[0021] Figure 3 A flowchart illustrating an information push method provided in an embodiment of this application;

[0022] Figure 4 A schematic diagram illustrating a scenario of an information push method provided in an embodiment of this application;

[0023] Figure 5 This is a schematic diagram of the structure of an information push device provided in an embodiment of this application;

[0024] Figure 6 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0025] In this embodiment, pets in the real world are referred to as physical pets, such as a first physical pet, a second physical pet, a target physical pet, etc. Pets in the network environment are referred to as virtual pets, or electronic pets, etc., such as a first virtual pet, a second virtual pet, a target virtual pet, etc. The object corresponding to a pet can be the pet's owner or caretaker, etc. The number of objects corresponding to a pet can be one, two, or more, and is not limited here.

[0026] This application does not limit the type of pet, and may include common pets such as cats, dogs, birds, and turtles, as well as less common pets such as chickens, pigs, snakes, lizards, geckos, and lions. In the embodiments of this application, pet information may include the pet's name, nose print, and other identifying information, as well as pet attributes such as breed, age, sex, skin color, and fur color. Furthermore, it may also include the pet's preference characteristics, such as dietary habits, sleep patterns, exercise preferences, and health status, which are not limited here.

[0027] This application pushes clothing recommendations for pets, and the pet information may also include the pet's body measurements. For example, if the pet is a dog, the pet information includes the dog's back length, chest circumference, and neck circumference. The back length is typically measured with a soft measuring tape along a straight line from the base of the neck to the base of the tail, and can be adjusted by approximately 3 centimeters. The chest circumference is typically measured with a soft measuring tape around the dog's chest along the top of its forelegs, and can be left 3-5 centimeters loose to minimize the amount of fur. The neck circumference is typically measured with a soft measuring tape around the dog's neck, and can be left 2-3 centimeters loose to minimize the amount of fur.

[0028] In this embodiment, pet clothing can include accessories such as clothes, hats, glasses, necklaces, and earrings. Clothing can include skirts, tops, pants, vests, etc., and can be categorized as trendy, traditional, fitted, or loose. Glasses can include over-ear glasses, colored contact lenses, lensless glasses, clear lenses, and tinted lenses. Necklaces and earrings can be categorized by material, such as metal accessories, yarn accessories, and beaded accessories.

[0029] It should be noted that there may be some differences between the pet information of physical pets and that of virtual pets. For example, the appearance of virtual pets may be interchangeable; virtual pets may react to various events, while physical pets may not be able to react to all of them; and the pet information of physical pets requires manual registration to be accessible, so there may be less reference data available for virtual pets.

[0030] In this embodiment, object information may include object attributes. These object attributes may include basic information such as the object's identification information (e.g., name, identity identifier, account identifier, etc.), age, gender, occupation, and address. They may also include the object's social data, such as social relationships online or in real life. Alternatively, object attributes may include tags for the object, such as interests and habits.

[0031] Object information can also be categorized by time of occurrence into real-time data and historical data. Real-time data can include relevant data from the user's current terminal, such as currently searched keywords and browsed content. Historical data can include the object's previous shopping records and browsing history from other user terminals. Historical data can be used to obtain the object's preference characteristics, such as their willingness to purchase pet clothing based on price, quality, brand, and effectiveness. Thus, based on whether the data characteristics of the real-time scenario match the object's preference characteristics, it can be determined whether to send a corresponding push notification to the object.

[0032] Pet information and object information can be uploaded to a server for storage or stored on a blockchain. The blockchain referred to in this application is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying platform, a platform product service layer, and an application service layer. Thus, by distributing data storage through the blockchain, data security can be ensured while enabling data sharing between different platforms.

[0033] This application does not limit the scenarios for storing pet information and object information; it can be used for scenarios such as pets visiting hospitals, receiving vaccinations, grooming, undergoing physical examinations, handling hospitalization procedures, handling insurance procedures, and obtaining identity information. It can be understood that after storing pet information and object information, information about the pet can be pushed to the target based on this information, improving the accuracy of the pushed pet information.

[0034] It should be noted that before acquiring data such as object information and pet information in the embodiments of this application, a prompt interface or pop-up window may be displayed. This prompt interface or pop-up window is used to prompt the user that the object information and the pet information fed by the object and the pet information fed by the object are being collected. The data acquisition steps will only begin after the user confirms the prompt interface or pop-up window; otherwise, the process will end.

[0035] It is understood that the specific implementation of this application may involve data of users, enterprises, institutions, etc. When the above embodiments of this application are applied to specific products or technologies, permission or consent from users, enterprises, institutions, etc. is required, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0036] 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 of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0037] Please see Figure 1 , Figure 1 This is a schematic diagram of a network architecture provided in an embodiment of this application. Figure 1 As shown, this network architecture may include a server 10d and a user terminal cluster. The user terminal cluster may include one or more user terminals; the number of user terminals is not limited here. Figure 1 As shown, the user terminal cluster may specifically include user terminal 10a, user terminal 10b, and user terminal 10c, etc.

[0038] Among them, server 10d can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0039] User terminals 10a, 10b, and 10c can all include: smartphones, tablets, laptops, PDAs, mobile internet devices (MIDs), wearable devices (such as smartwatches and smart bracelets), smart voice interaction devices, smart home appliances (such as smart TVs), and in-vehicle devices, as well as other electronic devices with video / image playback functions.

[0040] like Figure 1 As shown, user terminals 10a, 10b, and 10c can each connect to server 10d via a network, enabling data interaction between each user terminal and server 10d. For example, a first object sends its object information to server 10d through user terminal 10a, while a second object sends feedback on pet outfits to server 10d through user terminal 10b. Alternatively, server 10d can send a push notification about the pet's outfit to a target object through user terminal 10a.

[0041] The following example uses user terminal 10a. Please refer to [link / reference]. Figure 2 , Figure 2 This is a schematic diagram illustrating a scenario of purchasing clothing accessories for a physical pet, as provided in an embodiment of this application. Figure 2 As shown, the object 101 corresponding to the physical pet can open the shopping webpage 201 provided by the server 10d through the user terminal 10a. It can enter "pet clothing accessories" in the search component 202, or click on the component 203 for purchasing pet clothing accessories, etc., to generate a search request for physical pet clothing accessories through the user terminal 10a and submit the search request to the server 10d. After receiving the search request, the server 10d will obtain the physical pet's clothing push message and send it to the user terminal 10a, so that the page 204 corresponding to the clothing push message will be displayed on the user terminal 10a. If the object 101 selects component 205 of clothing accessory C on page 204 and completes the payment operation on the page corresponding to component 205... Figure 2 If (not shown in the image), then user terminal 10a selects accessory C as the target accessory and sends a delivery request for the target accessory to server 10d, so that server 10d notifies the merchant 102 corresponding to the target accessory. Merchant 102 will send a package 20 containing the target accessory to deliveryman 103, and deliveryman 103 can deliver the package 20 to object 101.

[0042] Object 101 can customize the physical pet with the desired outfit. If Object 101 is not satisfied with the desired outfit, it can submit a return request to Merchant 102 through User Terminal 10a. If Object 101 is satisfied with the desired outfit, it can submit a confirmation request to Merchant 102 through User Terminal 10a, or wait for Server 10d to automatically submit a confirmation request to Merchant 102. Object 101 can post comments about the desired outfit through User Terminal 10a. Figure 2 (Not shown in the image).

[0043] Reviews can be posted on the platform corresponding to server 10d, or on other platforms. Reviews can address various aspects such as price, brand, quality, how well the pet looks dressed, and the match between the actual item and the image, without any restrictions.

[0044] The process of purchasing outfits and accessories for virtual pets can be compared with... Figure 2 The content shown is consistent, or the user can receive the selection action from the virtual pet on the page corresponding to the outfit push message displayed on the user's terminal. Each selection action can generate a preview image of the outfit corresponding to the selected action. If the user is satisfied with the outfit, they can instruct the virtual pet to complete the payment or add it to their cart. It should be noted that the virtual pet may not notify the corresponding merchant and delivery person to deliver the outfit purchased for it. The virtual pet may also post comments on the purchased or unpurchased outfits on the server's platform or other platforms.

[0045] This application does not limit the scenario for the subject to choose clothing and accessories for their pet. Figure 2 The image illustrates a scenario where an object searches for clothing accessories. Alternatively, by analyzing a pet's growth data, if a new outfit is detected, a push notification can be sent to the object, allowing the object to select the desired outfit. Or, by analyzing an object's purchasing habits, if the object is likely interested in purchasing pet clothing accessories at a given time, a push notification can be sent to allow the object to select the desired outfit. Another approach is to distribute push notifications about pet clothing accessories in public places; if an object is interested in this data, further push notifications can be sent to that object, including information about the object and the pet it feeds.

[0046] The information push method provided in the embodiments of this application is described in detail below. Please refer to... Figure 3 , Figure 3 This is a flowchart illustrating an information push method provided in an embodiment of this application. The method can be executed by a computer device, which can be a server (e.g., Figure 1 The corresponding embodiment is a server 10d), or a user terminal (e.g., Figure 1 This could refer to any user terminal in the user terminal cluster shown, or a computer program (including program code), etc. Figure 3 As shown, the method includes the following steps S301 to S305, wherein:

[0047] Step S301: Obtain the pet information of the target entity pet and the object information of the target object corresponding to the target entity pet.

[0048] The information about the pet and the object can be referred to above or below. The target physical pet is the physical pet for which the pet outfit is to be pushed. The target object corresponding to the target physical pet can be the owner or caretaker of the target physical pet, etc., which will not be elaborated here.

[0049] This application does not limit the target physical pet and the target object. The target physical pet can be the physical pet fed by the target object currently searching for clothing. Alternatively, it can determine the physical pet that needs a new outfit based on the physical pet's growth data, such as whether it has grown taller, gained weight, or lost weight. In this case, the physical pet can be designated as the target physical pet, and the object feeding the target physical pet can be designated as the target object. Alternatively, it can determine the object's potential interest in purchasing pet-related clothing accessories based on their purchasing habits. For example, if the current time period is after payday, before a pet's birthday, or when the fixed duration of the clothing accessory is approaching, then the object can be identified as the target object, and the physical pet fed by that object can be designated as the target physical pet. Alternatively, it can determine the object as the target object and the pet fed by that object when feedback data is detected regarding push notifications for pet-related clothing accessories placed in public places is detected.

[0050] If the pet information of the target entity's pet and the object information of the target object are stored in the same location on the blockchain or server, then the pet information of the target entity's pet and the object information of the target object can be retrieved based on the target entity's pet identification information, such as name, nose print, pet number, etc. Alternatively, the object information of the target object and the pet information of the target entity's pet can be retrieved based on the target entity's pet identification information, and the object information of the target object can be retrieved based on the target object's identification information, within authorized limits.

[0051] It should be noted that the pet and object information above may change over time. Therefore, when obtaining pet and object information, the corresponding time information should also be obtained. This allows for the analysis of the characteristics and patterns of change in the pet and object information.

[0052] Step S302: Based on the pet information of the target entity pet and the object information of the target object, find the object cluster corresponding to the target object.

[0053] The target object cluster includes at least the target object, and may also include a first object corresponding to the first pet and a second object corresponding to the second pet. The first pet is any physical pet in the physical pet cluster corresponding to the target physical pet, and the second pet is any virtual pet in the virtual pet cluster corresponding to the target physical pet. The first and second pets can be understood as pets similar to the target physical pet in terms of species, appearance, age, gender, and living environment. The first and second objects can be understood as objects similar to the target object in terms of age, gender, consumption habits, clothing preferences, and hobbies. The first object is the owner or caretaker of the first pet, and the second object is the owner or caretaker of the second pet. The first and second objects can be understood as reference objects for the target object when purchasing pet clothing accessories, used to obtain clothing-related push notifications for the target physical pet.

[0054] This application does not limit the first object, the second object, the first pet, and the second pet, nor does it limit the method for finding the above objects and pets. In one possible example, step S302 may include the following steps: obtaining a first tag of the target entity pet based on the pet information of the target entity pet, and obtaining a second tag of the target object based on the object information of the target object; based on the first tag and the second tag, finding the entity pet cluster and virtual pet cluster corresponding to the target entity pet, and the object cluster corresponding to the target object.

[0055] The first tag describes the characteristics of the target pet. This allows for the search of the target pet or similar pets based on the first tag. The first tag can be described in key-value pair format, with the attribute type as the key and the attribute information as the value. Alternatively, the attribute information can be used alone as the first tag. The first tag can include characteristics such as the target pet's breed, age, gender, weight, and hobbies, without limitation. The number of first tags can be one, two, or more.

[0056] The second tag is used to describe the characteristics of the target object. Thus, the target object or similar objects can be found based on the second tag. The second tag can be described in the form of key-value pairs or as attribute information alone. The second tag may include characteristics of the first object such as age, gender, and hobbies, without limitation. The number of second tags can be one, two, or more.

[0057] It should be noted that when the number of first and second tags is greater than or equal to two, the weights of each first and second tag can be different. This allows for finding similar objects and pets by using weights, thus improving search accuracy.

[0058] Pets in both physical and virtual pet clusters must have at least one first tag, and objects in an object cluster must have at least one second tag. The following explanation uses a method for finding virtual pet clusters based on first tags as an example; physical pet clusters can be searched using the same method. First, find virtual pets that satisfy all first tags. If the number of found virtual pets is less than a threshold, you can select virtual pets to search based on a subset of first tags. For example, you can search a preset number of virtual pets in descending order of the weight of the first tags. Alternatively, you can determine the tags of pet clusters that satisfy all first tags, and then search based on those tags to find virtual pets similar to the target physical pet. This ensures both the quantity of pet clusters and the similarity between pets in a pet cluster and the target physical pet.

[0059] After finding a certain number of virtual pets, the target object's corresponding object cluster can be selected based on the matching value between the object's tag and the second tag. The virtual pets corresponding to the objects in the object cluster other than the target object can then be identified.

[0060] The above method involves first identifying a certain number of virtual pets, and then determining whether the objects corresponding to these virtual pets are the same as the target object. If so, the virtual pets fed by these objects are identified as virtual pets within the virtual pet cluster. Alternatively, a certain number of objects can be identified first, and then it can be determined whether the pets (physical and virtual pets) fed by these objects are the same as the target physical pet. If so, the objects corresponding to these pets are identified as the target object. Another approach is to identify a certain number of pets based on a first tag, and a certain number of objects based on a second tag, and then determine the physical pet cluster and virtual pet cluster corresponding to the target physical pet, as well as the object cluster corresponding to the target object, based on whether there is a relationship between these pets and objects. This improves the accuracy of obtaining the object clusters that the target object can reference, which is beneficial for improving the precision of pushing outfit-related messages.

[0061] Step S303: Obtain the outfit data set of the object cluster.

[0062] In this embodiment, the clothing data set of the object cluster includes a subset of clothing data for the target object, a subset of clothing data for the first object, and a subset of clothing data for the second object. Each subset of clothing data may include pet clothing data, feedback data, and scene data for each object regarding each piece of clothing data. Pet clothing data may include attributes of clothing accessories, such as type, material, comfort, and breathability. Feedback data may include comments from objects regarding clothing accessories, such as whether they like it, whether it fits well, and whether the price is reasonable. Feedback data may also include comments from other objects regarding clothing accessories, such as likes and comments from other objects on social media platforms. Scene data may include information such as the time and location of clothing accessories, or information such as the time and location of uploading to social media platforms.

[0063] This application does not limit the content of the clothing data set, nor does it limit the method of obtaining the clothing data set. It can retrieve pet clothing data from a blockchain or server based on the object's identification information and the pet's identification information, or it can send a data retrieval request to objects in an object cluster, causing the objects to respond with the corresponding clothing data set or fill in relevant information, etc.

[0064] Step S304: Based on the outfit data set, obtain the outfit push message for the target physical pet.

[0065] In this embodiment, the clothing recommendation message for a target pet is used to push clothing information of the target pet to a target object. The clothing recommendation message may include pet clothing reference data for the target clothing accessory. The target clothing accessory can be the clothing accessory of the target pet pushed to the target object. The pet clothing reference data for the target clothing accessory may include a reference clothing image of the target pet wearing the target clothing accessory, i.e., a preview image showing the clothing effect. Alternatively, it may include the recommendation message for the target clothing accessory, such as the pet clothing features of the target clothing accessory, or an event message about purchasing the target clothing accessory. This application does not limit the clothing recommendation message for the target pet or the method of obtaining the clothing recommendation message.

[0066] In one possible example, step S304 may include steps A1 to A4, wherein:

[0067] A1: Based on the clothing data set of the object cluster, obtain the pet clothing fusion features of the target object.

[0068] In this embodiment, the pet outfit fusion feature includes the features among the outfit features matched to the pet by objects in the object cluster, which can be understood as a weighted feature of the pet outfit features preferred by the target object. This application does not limit the method for obtaining the pet outfit fusion feature of the target object. It can determine the pet outfit preference features of the target object, the first object, and the second object respectively, and then retain the pet outfit preference features that are the same to obtain the pet outfit fusion feature of the target object.

[0069] In one possible example, step A1 may include the following steps A11 to A13, wherein:

[0070] A11: Based on the clothing data set, obtain the pet clothing preference features of the target object, the first object, and the second object.

[0071] Pet clothing preference characteristics refer to the pet clothing characteristics preferred by the target, such as color, type, fashionability, style, and length, etc., which are not limited here. The following method for obtaining the pet clothing preference characteristics of a target object is illustrated using pet clothing data for a first pet corresponding to the target object as an example. The pet clothing preference characteristics of the first and second objects can be obtained by referring to this method, and will not be elaborated further here. In one possible example, step A11 may include the following steps A111 to A114, wherein:

[0072] A111: Get the target object's preference value for the first pet's outfit.

[0073] A112: Based on the magnitude of the preference value, select at least two second pet outfits from multiple first pet outfits.

[0074] Here, the preference value refers to the target audience's liking for the first pet outfit. The first pet outfit refers to the styling of the accessories worn by the target audience on their physical pet or other pets (which can be physical or virtual). It should be understood that there can be multiple first pet outfits. This application does not limit the method for obtaining the preference value. It can be evaluated based on comments from the target audience's feedback data regarding the first pet outfit to obtain the target audience's satisfaction value for purchasing the first pet outfit. Alternatively, the target audience's preference value for the first pet outfit can be obtained based on factors such as the frequency of use of the accessories worn by the target audience when dressing their pet in the first pet outfit and / or the formality of the locations where these accessories appear, and the number of images and / or videos of the target audience wearing the accessories.

[0075] The second pet outfit is selected from the first pet outfits based on the highest preference value. This application does not limit the method for selecting the second pet outfit; the preference values ​​can be sorted in descending order, and the top N second pet outfits can be selected. N can be 5 or other fixed values. Alternatively, N can be based on the number of first pet outfits; for example, if the number of first pet outfits is 60, N can be 20% of 60, i.e., 12.

[0076] A113: Based on the pet outfit data of the second pet outfit, obtain the pet outfit preference sub-features of the second pet outfit.

[0077] The pet outfit preference sub-features of the second pet outfit can include data features of the pet outfit data, as well as the matching features between pet outfit accessories. These can be obtained by analyzing the data features of the pet outfit data and the matching features between pet outfit accessories.

[0078] A114: Based on the dressing weights of the second pet outfit and the pet dressing preference sub-features, obtain the pet dressing preference features of the target object.

[0079] The weight of the second pet outfit is used to limit the reliability of the pet outfit preference sub-feature of the second pet outfit. It can be determined based on the amount of feedback data from the target audience regarding the second pet outfit, the time interval between the time of the second pet outfit and the current time, or the sales volume of each accessory in the second pet outfit, etc., without limitation here. It can be understood that the more feedback data, the stronger the reference value of the pet outfit preference sub-feature of the second pet outfit. The longer the time interval, the weaker the reference value to the target audience. The higher the sales volume, the larger the number of people in the network who like the second pet outfit. Determining the weight of the second pet outfit based on one or more of the following factors—the amount of feedback data, the length of the time interval, and the sales volume—can improve the accuracy of obtaining the outfit weight, which is beneficial to improving the accuracy of obtaining pet outfit preference features.

[0080] The target audience's pet clothing preference features can be categorized based on various dimensions within these features. These dimensions can include quality, style, and aesthetics, among others, and are not limited here. The pet clothing preference features for each dimension are then obtained by weighting the pet clothing preference sub-features of one or more second pet outfits corresponding to a given dimension with the overall weight of that second pet outfit. If a dimension corresponds to one second pet outfit, the pet clothing preference feature for that dimension is the weighted value. If a dimension corresponds to multiple second pet outfits, the pet clothing preference feature for that dimension is the sum of the weighted values.

[0081] It is understandable that in steps A111 to A114, at least two second pet outfits are selected from multiple first pet outfits based on the target audience's preference value for the first pet outfit. Then, the pet outfit preference sub-features obtained from the pet outfit data of the second pet outfits are combined with the outfit weights of the second pet outfits to obtain the target audience's pet outfit preference features. In other words, analyzing the target audience's pet outfit preference features from the perspective of the preference sub-features and outfit weights of the pet outfits they prefer can improve the accuracy of obtaining pet outfit preference features, which is beneficial for improving the precision of pushing target pet outfits.

[0082] A12: Obtain the object weights of the target object, the first object, and the second object respectively.

[0083] Among them, object weights are used to limit the reference value of the pet's outfit. This can be determined based on the relationship between the first object and the target object, the relationship between the second object and the target object, etc. In one possible example, step A12 may include steps A121 to A124, wherein:

[0084] A121: Get the first number of the first object's outfit data subset, get the second number of the second object's outfit data subset, and get the third number of the target object's outfit data subset.

[0085] A122: Obtain the object weight of the target object based on the third, first, and second quantities.

[0086] The first quantity represents the number of data subsets related to the outfits of the first object, the second quantity represents the number of data subsets related to the outfits of the second object, and the third quantity represents the number of data subsets related to the outfits of the target object. The number of data subsets refers to the data set of an outfit combination that an object can use to dress up a pet. For example, hat A and clothes A constitute an outfit combination, hat A and clothes B constitute an outfit combination, hat A, clothes B, and earring C constitute an outfit component, etc.

[0087] The object weight of the target object can be equal to the ratio of the third quantity to the sum of the first, second, and third quantities. Alternatively, it can be equal to the ratio of the third quantity to the sum of the third and first quantities, and the ratio of the third quantity to the ratio between the third and second quantities, the sum, average, or maximum value of these two ratios, etc., which are not limited here.

[0088] A123: Get the first similarity value between the first object and the target object, and get the second similarity value between the second object and the target object.

[0089] The first similarity value describes the similarity between the first object and the target object, while the second similarity value describes the similarity between the second object and the target object. The following example illustrates the method for obtaining the first similarity value; the second similarity value can be obtained using the same method. Specifically, the object dimensions for comparison can be determined, such as age, education level, income and expenditure, hobbies, pet experience, etc. Then, the similarity sub-values ​​between the first object and the target object are determined for each object dimension, and the similarity sub-values ​​are then weighted and calculated based on the weights corresponding to the object dimensions.

[0090] A124: Based on the first similarity value and the second similarity value, as well as the object weight of the target object, obtain the object weights of the first object and the second object respectively.

[0091] The sum of the object weights of the first and second objects and the object weight of the target object is 1. For example, if the object weight of the target object is 0.6, then the sum of the object weights of the first and second objects is 1 - 0.6, which is 0.4. The first and second similarity values ​​can be assigned based on the difference between 1 and the object weight of the target object. For example, if the difference is 0.4 and the ratio between the first and second similarity values ​​is 0.33, then the first object can have a similarity value of 0.3, and the second object can have a similarity value of 0.1.

[0092] It is understandable that in steps A121 to A124, the object weight of the target object is first obtained based on the first quantity of the first object's clothing data subset, the second quantity of the second object's clothing data subset, and the third quantity of the target object's clothing data subset. Then, based on the first similarity value between the first object and the target object, the second similarity value between the second object and the target object, and the object weight of the target object, the object weights of the first object and the second object are obtained. In other words, determining the object weight of the target object first, and then obtaining the object weights of the first and second objects respectively based on the similarity values ​​between the first and second objects and the target object, can improve the accuracy of weight allocation and thus improve the accuracy of obtaining the pet clothing fusion features of the target object.

[0093] It should be noted that steps A121 and A122 may be executed before step A123, or after step A123, or simultaneously with step A123; no limitation is made here.

[0094] A13: Based on the pet clothing preference features and object weights of the target object, the pet clothing preference features and object weights of the first object, and the pet clothing preference features and object weights of the second object, obtain the pet clothing fusion features of the target object.

[0095] The pet outfit fusion feature for the target object can be obtained by referring to the method of obtaining the pet outfit preference feature for the target object. First, the pet outfit preference features of each object are classified according to each dimension of the pet outfit feature. Then, the pet outfit preference features of each object are weighted and obtained to obtain the pet outfit preference features of each dimension. Finally, the pet outfit preference features of multiple dimensions are combined to obtain the pet outfit fusion feature of the target object.

[0096] It is understandable that in steps A11 to A13, the pet clothing preference features and object weights of the target object, the first object, and the second object are first obtained respectively. Then, based on the object weights of the target object, the first object, and the second object, the pet clothing preference features of the target object, the first object, and the second object are weighted and calculated to obtain the pet clothing fusion feature of the target object. In this way, by obtaining the pet clothing fusion feature of the object cluster based on the object weight and pet clothing preference features of each object in the object cluster, the accuracy of obtaining the pet clothing fusion feature can be improved, which is beneficial to improving the accuracy of selecting the target pet clothing.

[0097] A2: Obtain the matching value between the pet outfit features and the pet outfit fusion features of each pet outfit in the set of pet outfits to be pushed.

[0098] The pet outfit set can include multiple pet outfits, and each pet outfit can include at least one accessory. The matching value between the pet outfit features and the pet outfit fusion features can be understood as the probability that the target audience likes the pet outfit and the probability that it is suitable for the target pet. The higher the matching value, the higher the success rate of pushing the outfit to the target audience. Pet outfit features and pet outfit fusion features can be categorized, and then features within the same category can be matched to obtain matching sub-values. Then, the matching sub-values ​​can be weighted based on the weights of each type of feature to obtain the matching value between the pet outfit features and the pet outfit fusion features.

[0099] A3: Based on the size of the matching value, select the target pet outfit from the pet outfit collection.

[0100] In this embodiment, the target pet outfit is the pet outfit with the largest matching value selected from the pet outfit set. It refers to the pet outfit to be pushed to the target object in the outfit push message, and may include at least one outfit accessory. This application does not limit the method for selecting the target pet outfit; the matching values ​​can be sorted in descending order, and the top M second pet outfits can be selected. M can be 10 or other fixed values. M can also be determined based on the settings of the user terminal's display page, such as the font, the size of the display components, or the number of messages to be pushed. Alternatively, M can be based on the number of pet outfits in the pet outfit set; for example, if the number of pet outfits is 100, M can be 10% of 100, i.e., 10.

[0101] A4: Get the outfit push message for the target pet.

[0102] This application does not limit the number of target pet outfits. If there are multiple outfit components or multiple target pet outfits, corresponding outfit push messages can be generated based on the multiple outfit components. The outfit push messages for each target pet outfit can be obtained. The outfit push messages for each target pet outfit can be pushed according to the matching value. The outfit push message may include the characteristics of the outfit components of the target pet outfit, as well as the matching characteristics between the outfit components.

[0103] It can be understood that in steps A1 to A4, the pet's outfit features for the target object are first obtained based on the outfit data set of the object cluster. Then, based on the matching value between the outfit features and the pet outfit fusion features of each pet outfit in the set to be pushed, the target pet outfit for the target entity is selected from the pet outfit set. Finally, the outfit push message for the target pet outfit is obtained. In this way, the accuracy of pushing pet outfits can be improved, which is beneficial to improving the push effect and matching effect.

[0104] Step S305: Send an outfit recommendation message to the target audience.

[0105] This application does not limit the steps for sending outfit push messages to the target object. In one possible example, the pet outfit data for the second pet outfit includes scenario data of the target object purchasing the second pet outfit. Before step S304, it may also include: determining the target object's pet outfit purchase scenario based on the scenario data. Thus, when the pet outfit purchase scenario arrives, step S304 is executed.

[0106] The scenario for purchasing pet clothing can be determined by the time the target user purchases pet clothing using their mobile device. Alternatively, it can be determined by the application client used by the target user to purchase the pet clothing. It should be noted that this pet clothing purchase scenario is a scope that may include multiple time points and multiple locations.

[0107] In this example, the scenario for purchasing pet clothing is determined based on the target audience's preferred second pet clothing scenario data. Then, when the pet clothing purchase scenario occurs, a push notification about the clothing is sent to the target audience. This further improves the accuracy of the targeting data, thus enhancing the effectiveness of the campaign.

[0108] exist Figure 3 The method described first locates the object cluster corresponding to the target object based on the pet information of the target physical pet and the object information of the target object corresponding to the target physical pet. Then, based on the outfit data set of the object cluster, it obtains the outfit push message for the target physical pet. The object cluster includes the target object, the first object corresponding to the first pet in the physical pet cluster corresponding to the target physical pet, and the second object corresponding to the second pet in the virtual pet cluster corresponding to the target physical pet. Thus, the outfit push information for the target physical pet references the target object's previous outfit data, as well as the outfit data of other objects feeding physical pets and virtual pets corresponding to the target object, improving the accuracy of pet outfit pushes. After sending the outfit push message to the target object, the target object can purchase the outfit accessories in the push message, which helps improve the overall outfit effect.

[0109] For example, when a target user brings their pet to a target hospital for a checkup, it's found that the pet's growth rate meets a preset target, and the time interval since the last purchase of pet clothing accessories is greater than a preset duration. Therefore, the target hospital's server can obtain the target user's object information and the target pet's pet information. Based on this information, it then locates the corresponding object cluster. Using the clothing data set of the object cluster, it retrieves a clothing recommendation message for the target pet and sends it to the target user. This informs the target user that they can purchase new clothing accessories for their pet at this time and can choose from the clothing recommendations in the recommendation message, improving the convenience and effectiveness of the selection process.

[0110] In one possible example, the target pet's outfit includes a first outfit component and a second outfit component. After step S304, the following steps may also be included: receiving a replacement request from the target object for the first outfit component; searching for a third outfit component corresponding to the first and second outfit components from a pre-stored outfit component library based on the pet outfit fusion features; obtaining a reference outfit image of the target entity pet based on the third outfit component and the second outfit component; and sending the reference outfit image to the target object.

[0111] The replacement request is used to request the replacement of the first outfit accessory. If no replacement request for the second outfit accessory is received, it indicates that the second outfit accessory is satisfactory to the target object. This application does not limit the number of the first and second outfit accessories; it should be understood that the number of the first outfit accessory can be 0, 1, 2, or more. The number of the second outfit accessory can also be 0, 1, 2, or more. Furthermore, the number of both the first and second outfit accessories cannot be 0 simultaneously. If the number of the first outfit accessory is 0, the method is not executed. If the number of the second outfit accessory is 0, the method does not consider the second outfit accessory.

[0112] The third outfit accessory is a replacement for the first outfit accessory. This application does not limit the number of third outfit accessories; it can be one, two, or more. It is understood that simultaneously pushing multiple replaceable third outfit accessories to the target object is beneficial to improving the efficiency of the push.

[0113] The outfit accessory library can include outfit accessories from the aforementioned set of pet outfits to be pushed, and can also include other outfit accessories. The method for finding a third outfit accessory can include the following steps: searching the pre-stored outfit accessory library for a reference outfit accessory that matches the accessory type of the first outfit accessory and satisfies the pet outfit fusion characteristics; then obtaining the matching value between the reference outfit accessory and the second outfit accessory, and selecting the third outfit accessory from the reference outfit accessory based on this large configuration. In this way, the found third outfit accessory matches the type of the first outfit accessory, can be matched with the second outfit accessory, and satisfies the pet outfit fusion characteristics, improving the accuracy of finding and replacing outfit accessories and thus improving push efficiency.

[0114] Reference outfit images refer to the preview images displayed when the second and third outfit pieces are paired with the target pet. Based on the pet image in the target pet's pet information, images corresponding to the suitable second and third outfit pieces can be loaded. Alternatively, pairing suggestions for the second and third outfit pieces can be provided to help users combine these pieces with other outfits, thus improving the overall look.

[0115] For example, please refer to Figure 4, Figure 4 This is a schematic diagram illustrating a scenario for an information push method provided in this application. For example... Figure 4 As shown, target object 101 can select outfit A as the outfit to be replaced from the display list of target pet outfits in user terminal 10a, i.e., the first outfit. Then, the outfit B that target object 101 did not select is the second outfit. For example, the first outfit is clothes, and the second outfit is glasses 402. Then, a replacement request for the first outfit is sent to server 10d. Server 10d can search for the third outfit corresponding to the first outfit from the pre-stored outfit library based on the pet outfit fusion features, for example, clothes 401, and obtain a reference outfit image 400 (clothes 401 and glasses 402) of the target entity pet 40. Then, the reference outfit image 400 is sent to user terminal 10a so that target object 101 receives the reference outfit image 400. The target entity pet 40 in the reference outfit image 400 is displayed from an angle that is conducive to showing the characteristics of the outfit. In fact, the outfit push message can be displayed using a reference preview image of the target entity pet. Figure 4 (Not shown).

[0116] Understandably, in this example, if the target user is dissatisfied with the first outfit accessory in the target pet's outfit, they can send a replacement request to the information push device. The device then uses the pet's outfit fusion features to search for a third outfit accessory corresponding to the first and second outfit accessories in a pre-stored outfit accessory library. It then sends the target user a reference outfit image of the target pet obtained based on the third and second outfit accessories. In this way, the target user can determine whether to purchase the outfit accessory by comparing the replacement and non-replaced outfit images, thus improving push efficiency.

[0117] It should be noted that this application can also be applied to sending outfit recommendations for virtual pets to objects, as described in the method for sending outfit recommendations for target physical pets to target objects. For example, the process involves: obtaining the virtual pet's pet information and the object information corresponding to that virtual pet; searching for the object cluster corresponding to that object based on the virtual pet's pet information and the object information; obtaining the outfit data set of the object cluster; obtaining the virtual pet's outfit recommendation message based on the outfit data set; and sending the outfit recommendation message to the object. The object cluster includes the object to be recommended, the objects corresponding to physical pets in the physical pet cluster corresponding to the virtual pet, and the objects corresponding to virtual pets in the virtual pet cluster corresponding to the virtual pet. Thus, the outfit recommendation message sent to the object references the object's previous outfit data, as well as the outfit data of other objects that feed physical pets and objects that feed virtual pets, improving the accuracy of pet outfit recommendations and enhancing the overall matching effect.

[0118] The methods of the embodiments of this application have been described in detail above, and the apparatus of the embodiments of this application is provided below.

[0119] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of an information push device provided in an embodiment of this application. Figure 5 As shown, the information push device includes a processing unit 501 and a communication unit 502, wherein:

[0120] The processing unit 501 is used to obtain the pet information of the target entity pet and the object information of the target object corresponding to the target entity pet; based on the pet information of the target entity pet and the object information of the target object, it searches for the object cluster corresponding to the target object; the object cluster includes the first object corresponding to the first pet in the entity pet cluster corresponding to the target entity pet, the second object corresponding to the second pet in the virtual pet cluster corresponding to the target entity pet, and the target object; obtain the outfit data set of the object cluster; and based on the outfit data set, obtain the outfit push message of the target entity pet.

[0121] The communication unit 502 is used to send outfit recommendation messages to the target object.

[0122] In one possible example, the processing unit 501 is specifically used to obtain the pet outfit fusion features of the target object based on the outfit data set of the object cluster; obtain the matching value between the pet outfit features and pet outfit fusion features of each pet outfit in the pet outfit set to be pushed; select the target pet outfit of the target entity pet from the pet outfit set based on the size of the matching value; and obtain the outfit push message of the target pet outfit.

[0123] In one possible example, the processing unit 501 is specifically used to obtain the pet clothing preference features of the target object, the first object, and the second object respectively based on the clothing data set; obtain the object weights of the target object, the first object, and the second object respectively; and obtain the pet clothing fusion feature of the target object based on the pet clothing preference features and object weights of the target object, the pet clothing preference features and object weights of the first object, and the pet clothing preference features and object weights of the second object.

[0124] In one possible example, the outfit dataset includes pet outfit data for each of a plurality of first pet outfits for the target object; the processing unit 501 is specifically used to obtain the target object's preference value for the first pet outfit; select at least two second pet outfits from the plurality of first pet outfits based on the magnitude of the preference value; obtain pet outfit preference sub-features for the second pet outfits based on the pet outfit data of the second pet outfits; and obtain the target object's pet outfit preference features based on the outfit weights of the second pet outfits and the pet outfit preference sub-features.

[0125] In one possible example, the pet outfit data for the second pet outfit includes scenario data of the target object purchasing the second pet outfit; the processing unit 501 is also used to determine the target object's pet outfit purchase scenario based on the scenario data; the communication unit 502 is also used to send an outfit push message to the target object when the pet outfit purchase scenario arrives.

[0126] In one possible example, the processing unit 501 is specifically used to obtain a first quantity of the clothing data subset of the first object, a second quantity of the clothing data subset of the second object, and a third quantity of the clothing data subset of the target object; based on the third quantity, the first quantity, and the second quantity, obtain the object weight of the target object; obtain a first similarity value between the first object and the target object, and obtain a second similarity value between the second object and the target object; based on the first similarity value, the second similarity value, and the object weight of the target object, obtain the object weights of the first object and the second object respectively.

[0127] In one possible example, the target pet's outfit includes a first outfit component and a second outfit component; the communication unit 502 is also used to receive a replacement request from the target object for the first outfit component; the processing unit 501 is also used to search for a third outfit component corresponding to the first and second outfit components from a pre-stored outfit component library based on the pet outfit fusion features; based on the third outfit component and the second outfit component, obtain a reference outfit image of the target entity pet; the communication unit 502 is also used to send the reference outfit image to the target object.

[0128] Please refer to Figure 6 , Figure 6This is a schematic diagram of a computer device provided in an embodiment of this application. The computer device 600 includes a processor 601, a communication interface 602, and a memory 603. The processor 601, the communication interface 602, and the memory 603 can be interconnected via a bus 605 or other means. Figure 5 The functions implemented by the processing unit 501 shown can be implemented by one or more processors 601. Figure 5 The functions implemented by the communication unit 502 shown can be realized through the communication interface 602.

[0129] The processor 601 includes one or more processors, such as one or more central processing units (CPUs). When the processor 601 is a CPU, the CPU can be a single-core CPU or a multi-core CPU. In this embodiment, the processor 601 is used to control the computer device 600. Figure 3 The example shown.

[0130] The communication interface 602 is used to enable communication with other devices. For example, if the computer device 600 is a user terminal, the communication interface 602 can enable communication between the user terminal and devices such as servers; if the computer device 600 is a server, the communication interface 602 can enable communication between the server and devices such as user terminals.

[0131] The memory 603 includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or compact disc read-only memory (CD-ROM), and is used to store related instructions and data.

[0132] In this embodiment, memory 603 stores computer program 604, which includes program instructions. Processor 601 is configured to call the program instructions. The program includes instructions for performing the following steps:

[0133] Obtain the pet information of the target entity pet and the object information of the target object corresponding to the target entity pet;

[0134] Based on the pet information of the target entity pet and the object information of the target object, find the object cluster corresponding to the target object; the object cluster includes the first object corresponding to the first pet in the entity pet cluster corresponding to the target entity pet, the second object corresponding to the second pet in the virtual pet cluster corresponding to the target entity pet, and the target object;

[0135] Retrieve the outfit data set of the object cluster;

[0136] Based on the clothing data set, obtain clothing-related push messages for the target physical pet;

[0137] Send outfit recommendations to the target audience.

[0138] In a possible example, regarding retrieving the outfit-related push notification for a target entity pet based on an outfit data set, the instructions in the program are specifically used to perform the following operations:

[0139] Based on the clothing data set of object clusters, obtain the pet clothing fusion features of the target object;

[0140] Obtain the matching value between the pet outfit features and the pet outfit fusion features of each pet outfit in the set of pet outfits to be pushed;

[0141] Based on the size of the matching value, select the target pet outfit from the pet outfit collection;

[0142] Get outfit-related push notifications for the target pet.

[0143] In a possible example, regarding obtaining the pet's clothing blending features based on a clothing dataset, the instructions in the program are specifically used to perform the following operations:

[0144] Based on the clothing data set, we obtained the pet clothing preference features of the target object, the first object, and the second object respectively;

[0145] Obtain the object weights of the target object, the first object, and the second object respectively;

[0146] Based on the pet clothing preference features and object weights of the target object, the pet clothing preference features and object weights of the first object, and the pet clothing preference features and object weights of the second object, the pet clothing fusion features of the target object are obtained.

[0147] In one possible example, the outfit dataset includes pet outfit data for the target object for each of multiple first pet outfits; regarding the acquisition of pet outfit preference features of the target object, the first object, and the second object based on the outfit dataset, the instructions in the program are specifically used to perform the following operations:

[0148] Get the target user's preference value for the first pet's outfit;

[0149] Based on the preference value, select at least two second pet outfits from multiple first pet outfits;

[0150] Based on the pet outfit data of the second pet outfit, obtain the pet outfit preference sub-features of the second pet outfit;

[0151] Based on the dressing weights of the second pet outfit and the pet dressing preference sub-features, the pet dressing preference features of the target object are obtained.

[0152] In one possible example, the pet outfit data for the second pet outfit includes scenario data of the target audience purchasing the second pet outfit; the instructions in the program are also used to perform the following operations:

[0153] Based on scenario data, determine the target audience's purchase scenarios for pet clothing;

[0154] When the scene for purchasing pet outfits arrives, send an outfit recommendation message to the target audience.

[0155] In a possible example, the instructions in the program are specifically used to perform the following operations in order to obtain the object weights of the target object, the first object, and the second object respectively:

[0156] Get the first number of the first object's outfit data subset, get the second number of the second object's outfit data subset, and get the third number of the target object's outfit data subset;

[0157] Based on the third, first, and second quantities, obtain the object weight of the target object;

[0158] Get the first similarity value between the first object and the target object, and get the second similarity value between the second object and the target object;

[0159] Based on the first similarity value and the second similarity value, as well as the object weight of the target object, the object weights of the first object and the second object are obtained respectively.

[0160] In one possible example, the target pet's outfit includes a first outfit piece and a second outfit piece, and the instructions in the program are also used to perform the following operations:

[0161] Receive the target object's request to replace the first accessory;

[0162] Based on the pet outfit fusion feature, the third outfit corresponding to the first and second outfits is searched from the pre-stored outfit accessory library;

[0163] Based on the third and second outfit accessories, obtain the reference outfit image of the target physical pet;

[0164] Send reference outfit images to the target audience.

[0165] This application also provides a computer-readable storage medium storing a computer program executed by the aforementioned information push device. The computer program includes program instructions, which, when executed by a processor, enable the execution of the aforementioned... Figure 3 The description of the information push method in the corresponding embodiments is already provided and will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the computer-readable storage medium embodiments related to this application, please refer to the description of the method embodiments of this application. As an example, program instructions can be deployed and executed on a single computing device, or on multiple computing devices located in one location, or on multiple computing devices distributed across multiple locations and interconnected via a communication network. These multiple computing devices distributed across multiple locations and interconnected via a communication network can constitute a blockchain system.

[0166] Furthermore, it should be noted that this application also provides a computer program product or computer program, which may include computer instructions, which may be stored in a computer-readable storage medium. The processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor may execute the computer instructions, causing the computer device to perform the aforementioned actions. Figure 3 The description of the information push method in the corresponding embodiments is already provided, and therefore will not be repeated here. Furthermore, the beneficial effects of using the same method will also not be repeated. For technical details not disclosed in the computer program products or computer program embodiments involved in this application, please refer to the description of the method embodiments of this application.

[0167] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0168] The steps in the method of this application embodiment can be adjusted, combined, or deleted according to actual needs.

[0169] The modules in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0170] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. The storage medium can be a magnetic disk, optical disk, ROM, or RAM, etc.

[0171] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. An information push method, characterized in that, include: Obtain the pet information of the target entity pet and the object information of the target object corresponding to the target entity pet; Based on the pet information of the target entity pet and the object information of the target object, the object cluster corresponding to the target object is searched; the object cluster includes the first object corresponding to the first pet in the entity pet cluster corresponding to the target entity pet, the second object corresponding to the second pet in the virtual pet cluster corresponding to the target entity pet, and the target object; Obtain the outfit data set of the object cluster. The outfit data set of the object cluster includes the outfit data subset of the target object, the outfit data subset of the first object, and the outfit data subset of the second object. Each outfit data subset includes pet outfit data, feedback data, and scene data for each object in relation to each outfit data. Based on the outfit data set, obtain the outfit push message for the target entity pet; Send the outfit recommendation message to the target object.

2. The method as described in claim 1, characterized in that, The step of obtaining the outfit push message for the target entity pet based on the outfit data set includes: Based on the outfit data set of the object cluster, the pet outfit fusion feature of the target object is obtained. The pet outfit fusion feature includes the features between the outfit features of the objects in the object cluster that are matched with the pet. Obtain the matching value between the pet outfit features of each pet outfit in the set of pet outfits to be pushed and the pet outfit fusion features; Based on the magnitude of the matching value, select the target pet outfit for the target entity pet from the pet outfit set; Obtain the outfit push message for the target pet; The step of obtaining the pet's outfit fusion features based on the outfit data set includes: Based on the clothing data set, the pet clothing preference features of the target object, the first object, and the second object are obtained respectively; Obtain the object weights of the target object, the first object, and the second object respectively; Based on the pet outfit preference features and object weights of the target object, the pet outfit preference features and object weights of the first object, and the pet outfit preference features and object weights of the second object, the pet outfit fusion features of the target object are obtained.

3. The method as described in claim 2, characterized in that, The outfit data set includes pet outfit data for each of the multiple first pet outfits for the target object; The step of obtaining the pet clothing preference features of the target object, the first object, and the second object based on the clothing data set includes: Obtain the target object's preference value for the first pet's outfit; Based on the magnitude of the preference value, at least two second pet outfits are selected from the plurality of first pet outfits; Based on the pet outfit data of the second pet outfit, obtain the pet outfit preference sub-features of the second pet outfit; Based on the dressing weight of the second pet outfit and the pet dressing preference sub-feature, the pet dressing preference features of the target object are obtained.

4. The method as described in claim 3, characterized in that, The pet outfit data for the second pet outfit includes scenario data of the target object purchasing the second pet outfit; the method further includes: Based on the scenario data, determine the target object's pet clothing purchase scenario; When the scenario of purchasing pet clothing arrives, a push notification message about the clothing is sent to the target object.

5. The method as described in claim 2, characterized in that, The step of obtaining the object weights of the target object, the first object, and the second object respectively includes: Obtain the first number of the first object's outfit data subset, obtain the second number of the second object's outfit data subset, and obtain the third number of the target object's outfit data subset; Based on the third quantity, the first quantity, and the second quantity, the object weight of the target object is obtained; Obtain a first similarity value between the first object and the target object, and obtain a second similarity value between the second object and the target object; Based on the first similarity value and the second similarity value, as well as the object weight of the target object, the object weights of the first object and the second object are obtained respectively.

6. The method according to any one of claims 1-5, characterized in that, The target pet outfit includes a first outfit component and a second outfit component; the method further includes: Receive the target object's request to replace the first clothing accessory; Based on the pet outfit fusion feature, the third outfit corresponding to the first outfit and the second outfit is searched from the pre-stored outfit accessory library; Based on the third and second outfit accessories, obtain a reference outfit image of the target physical pet; The reference outfit image is sent to the target object.

7. An information push device, characterized in that, Includes units for performing the method as described in any one of claims 1-6.

8. A computer device, characterized in that, It includes a memory and a processor; the memory is connected to the processor, the memory is used to store a computer program, and the processor is used to invoke the computer program so that the computer device performs the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program adapted to be loaded and executed by a processor to cause a computer device having the processor to perform the method of any one of claims 1-6.

Images