Information pushing method, device, equipment and medium

Abstract

The information pushing method, device, equipment and medium provided by the application include: obtaining a recommendation request including target commodity information; using the target commodity information to search a list database to determine a target recommendation list corresponding to the target commodity information; wherein the list database is obtained based on full commodity information in a previous calculation period; obtaining relationship graph data of newly added commodity information added after the previous calculation period and part of the commodity information in the full commodity information; when the relationship graph data contains commodity information in the target recommendation list, using the relationship graph data to update the target recommendation list corresponding to the target commodity information. The application solves the problem that after a newly added commodity goes online, a user cannot obtain the newly added commodity in a recommendation list in time, resulting in low accuracy of the recommended result in the recommendation list and poor recommendation effect of the newly added commodity.

Description

Technical Field

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

[0002] With the development of internet technology, shopping on e-commerce platforms has become increasingly popular. Compared to traditional brick-and-mortar stores, the advantage of e-commerce platforms is that they can recommend relevant products to users, who can then directly purchase items from the recommended list, thus improving shopping efficiency.

[0003] The current information push method generates a list database offline based on the feature information of all user information and all product information within a fixed calculation cycle. Based on this list database, a target recommendation list is determined. When a new product is added, the server only calculates the related recommendation list for the new product or displays a default recommendation list within the related recommendations. The new product's target recommendation list is only updated when the server generates the list database offline in the next calculation cycle.

[0004] In other words, after a new product is launched, users cannot see it in the recommendation list in a timely manner, nor can they purchase it based on the recommendation list. This reduces the accuracy of the recommendation results and diminishes the effectiveness of recommending the new product. Summary of the Invention

[0005] This application provides an information push method, apparatus, device, and medium to solve the problem that after new products are launched, users cannot obtain the new products in the recommendation list in a timely manner, nor can they purchase the new products based on the recommendation list, resulting in low accuracy of the recommendation list results and poor recommendation effect of new products.

[0006] In a first aspect, this application provides an information push method, the method being applied to a server, the method comprising:

[0007] Obtain a recommendation request; wherein the recommendation request includes target product information;

[0008] The target product information is used to search the list database to determine the target recommendation list corresponding to the target product information; wherein, the list database was obtained based on all product information in the previous calculation cycle;

[0009] Obtain the relationship graph data between the newly added product information added after the previous calculation cycle and the partial product information in the full product information;

[0010] When the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information and a portion of the product information from the full product information.

[0011] In one specific implementation, before acquiring the relationship graph data between the newly added product information added after the previous calculation cycle and the partial product information in the total product information, the method further includes:

[0012] Select the associated product information and the similar product information corresponding to the newly added product information from the full product information;

[0013] The relationship graph data is constructed based on the similarity between the similar product information and the newly added product information, and the number of users corresponding to the associated product information.

[0014] In one specific implementation, selecting associated product information and similar product information corresponding to the newly added product information from the full product information specifically includes:

[0015] Obtain the feature information of newly added product information, and calculate and determine the similarity between the feature information of the newly added product information and the feature information of each product information in the full product information;

[0016] Product information with a similarity greater than a preset threshold is selected from the full product information as the similar product information;

[0017] The similar related product information is used to search the list database to determine the recommendation list corresponding to the similar product information;

[0018] The product information in the recommendation list corresponding to the similar related product information is used as the related product information.

[0019] In one specific implementation, the relationship graph data is constructed based on the similarity between the similar product information and the newly added product information and the number of users corresponding to the associated product information, specifically including:

[0020] The similar product information, the related product information, and the newly added product information are used as nodes in the relationship graph data;

[0021] The similarity between the similar product information and the newly added product information is used as the edge length between the second node corresponding to the similar product information and the first node corresponding to the newly added product information.

[0022] The number of users corresponding to the associated product information is used as the edge length between the second node corresponding to the similar product information and the third node corresponding to the associated product information.

[0023] In one specific implementation, when the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information. This specifically includes:

[0024] For each baseline product information, the correlation between the new product information and the baseline product information is calculated based on the path length and path weight between the fourth node corresponding to the baseline product information and the first node corresponding to the new product information.

[0025] Based on the correlation between each baseline product information and the newly added product information, the position of the newly added product information in the target recommendation list is determined, and the newly added product information is inserted into the target recommendation list according to its position in the target recommendation list;

[0026] The benchmark product information refers to product information that exists both in the target recommendation list and in the relationship graph data.

[0027] In one specific embodiment, before using the target product information lookup list database to determine the target recommendation list corresponding to the target product information, the method further includes:

[0028] When the calculation cycle is reached, acquire the feature information of all user information and the feature information of all product information;

[0029] The list database is updated based on the feature information of all user information and the feature information of all product information.

[0030] In one specific embodiment, the method further includes:

[0031] Send the updated list of recommended targets to the client.

[0032] Secondly, this application provides an information push device, the device comprising:

[0033] An acquisition module is used to acquire recommendation requests; wherein, the recommendation request includes target product information;

[0034] The processing module is used to search the list database using the target product information to determine the target recommendation list corresponding to the target product information; wherein, the list database is obtained based on full product information in the previous calculation cycle;

[0035] The acquisition module is also used to acquire the relationship graph data between the newly added product information added after the previous calculation cycle and the partial product information in the full product information;

[0036] The processing module is further configured to update the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and the partial product information in the full product information when the relationship graph data contains product information in the target recommendation list.

[0037] Thirdly, this application provides an electronic device, comprising:

[0038] Processor, memory;

[0039] The memory is used to store the executable instructions of the processor;

[0040] The processor is configured to execute the information push method described in the first aspect by executing the executable instructions.

[0041] Fourthly, this application provides a readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the information push method described in the first aspect.

[0042] This application provides an information push method, apparatus, device, and medium. The information push method provided by this application is executed through the following steps: the server obtains a recommendation request including target product information; the target product information is used to search the list database obtained based on the full product information in the previous calculation cycle to determine the target recommendation list corresponding to the target product information; relationship graph data of newly added product information and partial product information in the full product information is obtained after the previous calculation cycle; when the relationship graph data contains product information in the target recommendation list, it indicates that there are related or similar products of the newly added product in the target recommendation list, and the target recommendation list corresponding to the target product information is updated according to the relationship graph data. Compared to adding newly added product information to the full product information in real time to calculate the global recommendation list, this application simplifies the computation, shortens the computation time, and solves the problem that after a new product is launched, users cannot obtain the new product in the recommendation list in a timely manner, nor can they purchase the new product based on the recommendation list, resulting in low accuracy of recommendation results and poor recommendation effect of new products. Attached Figure Description

[0043] 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0044] Figure 1 A schematic diagram illustrating the application scenario of the information push method provided in this application;

[0045] Figure 2 A flowchart illustrating an embodiment of the information push method provided in this application;

[0046] Figure 3 A flowchart illustrating Embodiment 2 of the information push method provided in this application;

[0047] Figure 4 A schematic diagram of the structure of an embodiment of the information push device provided in this application;

[0048] Figure 5 A schematic diagram of the structure of an electronic device provided in this application;

[0049] Figure 6 A schematic diagram of the information push method provided in this application;

[0050] Figure 7 A schematic diagram illustrating the information push method provided in this application. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, 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, not all embodiments. Based on the embodiments of this application, all other embodiments made by those skilled in the art under the guidance of these embodiments are within the scope of protection of this application.

[0052] The terms “first,” “second,” “third,” “fourth,” etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a particular order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms “comprising” and “having,” and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0053] With the development of internet technology, shopping on e-commerce platforms has become increasingly popular. Compared to traditional brick-and-mortar stores, e-commerce platforms have the advantage of using recommendation algorithms to calculate product recommendation lists and recommend relevant products to users. Users can directly purchase recommended products from these lists, improving shopping efficiency. Currently, the information push method generates a list database offline based on the feature information of all user information and all product information within a fixed calculation cycle, determining the target recommendation list. When a new product is added, the server only calculates the relevant recommendation list for the new product or displays a default recommendation list within the relevant recommendations. The new product is only updated in the user's target recommendation list when the server generates the list database offline in the next calculation cycle. In other words, after a new product is added, users cannot immediately see it in the recommendation list or purchase it based on the list, reducing the accuracy of the recommendation results and the effectiveness of the new product recommendation.

[0054] To address the problems existing in existing technologies, the inventors, during their research on information push methods, discovered that by using similar product information, related product information, and newly added product information as nodes in a relationship graph data set, and using the similarity between similar and newly added product information as the edge length between the second node corresponding to similar product information and the first node corresponding to newly added product information, and using the number of users corresponding to related product information as the edge length between the second node corresponding to similar product information and the third node corresponding to related product information, relationship graph data can be quickly constructed, improving the efficiency of relationship graph data construction. When a recommendation request containing target product information is received from a client, the server uses the target product information to search the list database obtained based on all product information in the previous calculation cycle to determine the target recommendation list corresponding to the target product information; it then obtains the relationship graph data of newly added product information and some product information from the all product information added after the previous calculation cycle; when the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information and some product information from the all product information. Based on this method, it is possible to update the user's recommendation list for newly added products without waiting for the server to generate the list database offline in the next calculation cycle. This invention addresses the problem that after new products are added, users cannot promptly see them in the recommendation list, nor can they purchase them based on the recommendation list, leading to low accuracy and poor performance in recommending new products. Based on the above inventive concept, the information push scheme described in this application was designed.

[0055] The information push scheme of this application is described in detail below.

[0056] Figure 1 This application of the information push method is illustrated in the following diagram: Figure 1 As shown, this application scenario can include: client 101 and server 102.

[0057] For example, in Figure 1In the illustrated application scenario, when a user wants to obtain a recommendation list, client 101 sends a recommendation request to server 102. The recommendation request includes target product information. After receiving the request, server 102 uses the target product information to search the list database to determine the target recommendation list corresponding to the target product information. The list database is obtained by server 102 based on all product information in the previous calculation cycle. Server 102 also obtains relationship graph data of newly added product information and some product information from the full product information after the previous calculation cycle. When the relationship graph data contains product information from the target recommendation list, server 102 updates the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and some product information from the full product information. Server 102 then sends the updated target recommendation list to client 101.

[0058] Based on the above process, when a user has a recommendation request, they can directly send a recommendation request, including target product information, to the server through the client. The server uses the target product information to search the list database obtained from the full product information in the previous calculation cycle to determine the target recommendation list corresponding to the target product information. The server obtains the relationship graph data of newly added product information and some product information from the full product information after the previous calculation cycle. When the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information and some product information from the full product information. In other words, after a new product is launched, when a user needs to obtain the recommendation list, the user receives the updated recommendation list. This application solves the problem that after a new product is launched, users cannot obtain the new product in the recommendation list in a timely manner, resulting in low accuracy of recommendation results and poor recommendation effect of new products.

[0059] It should be noted that, Figure 1 This is merely a schematic diagram illustrating one application scenario provided by an embodiment of this application. This embodiment does not necessarily represent... Figure 1 The document does not limit the actual form of the various devices included, nor does it specify the form of the devices. Figure 1 The interaction methods between devices are limited, and can be set according to actual needs in the specific application of the solution.

[0060] The technical solution of this application will now be described in detail through specific embodiments. It should be noted that the following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0061] Figure 2 A flowchart illustrating an embodiment of the information push method provided in this application. See also... Figure 2 The information push method specifically includes the following steps:

[0062] Step S201: Obtain the recommendation request.

[0063] The recommendation request includes information about the target product.

[0064] When a user needs to retrieve a recommendation list, the user enters the target product information on the client. The client then generates a recommendation request in real time based on the user's input and sends it to the server. The server receives the recommendation request from the client.

[0065] In one specific implementation, the client displays a visual interface that shows multiple selectable parameter information and a search bar.

[0066] Optionally, the user determines the required parameters based on the product recommendation request. The client can obtain the recommendation requirement information based on the parameter information selected by the user, and the client sends a recommendation request to the server, which includes the target product information.

[0067] Optionally, the user enters target product information in the search bar based on the product recommendation request. The client can obtain recommendation request information based on the target product information entered by the user, and the client sends a recommendation request to the server, which includes the target product information.

[0068] Step S202: Use the target product information lookup list database to determine the target recommendation list corresponding to the target product information.

[0069] The list database includes multiple recommendation lists. Figure 6 A schematic diagram of the information push method provided in this application, such as Figure 6 As shown, when the server receives a recommendation request, it determines the target recommendation list corresponding to the target product information from multiple recommendation lists included in the list database based on the target product information included in the recommendation request.

[0070] The list database was obtained in the previous calculation cycle based on full product information.

[0071] Specifically, when the calculation cycle is reached, the server obtains the feature information of all user information and the feature information of all product information;

[0072] The server updates the list database based on the feature information of all user information and the feature information of all product information.

[0073] Optionally, the server obtains feature information of all user information and feature information of all product information, calculates a recommendation list using a collaborative filtering algorithm, and updates the list database.

[0074] Optionally, the server obtains feature information of all user information and feature information of all product information, uses Factorization Machines (FM) to calculate a recommendation list, and updates the list database.

[0075] Step S203: Obtain the relationship graph data of newly added product information and partial product information in the full product information after the previous calculation cycle.

[0076] New products are added after the end of the previous calculation cycle and before the server receives recommendation requests from clients. Since these new products were not added when the server updated the list database based on the feature information of all user information and all product information in the previous calculation cycle, the list database calculated in the previous cycle did not contain any new product information. In other words, the target recommendation list does not include new product information. To update the target recommendation list to include new product information, the server retrieves the relationship graph data between the new product information added after the previous calculation cycle and a portion of the product information from the total product information.

[0077] Step S204: When the relationship graph data contains product information from the target recommendation list, update the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and a portion of the product information from the full product information.

[0078] When the relationship graph data contains product information from the target recommendation list, it means that the target recommendation list includes related products or similar products of the newly added product. This product information, which exists in both the target recommendation list and the relationship graph data, serves as the baseline product information.

[0079] For each baseline product information, the correlation between the new product information and the baseline product information is calculated based on the path length and path weight between the fourth node corresponding to the baseline product information and the first node corresponding to the new product information. Based on the correlation between each baseline product information and the new product information, the position of the new product information in the target recommendation list is determined, and the new product information is inserted into the target recommendation list according to its position.

[0080] Figure 6 A schematic diagram of the information push method provided in this application, such as Figure 6 As shown, optionally, when the relationship graph data contains product information from the target recommendation list, after updating the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and some product information from the full product information, the server sends the updated target recommendation list to the client.

[0081] It should be noted that, as Figure 6 As shown, when the relationship graph data does not contain product information from the target recommendation list, there is no need to update the target recommendation list corresponding to the target product information.

[0082] The information push method provided in this embodiment is executed through the following steps: obtaining a recommendation request; using the target product information to search the list database to determine the target recommendation list corresponding to the target product information; obtaining the relationship graph data of newly added product information and partial product information from the full product information after the previous calculation cycle; when the relationship graph data contains product information from the target recommendation list, updating the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and partial product information from the full product information. Compared with the prior art, in this embodiment, when a user has a recommendation request, the server obtains the recommendation request and uses the target product information included in the recommendation request to search the list database generated in the previous calculation cycle for the target recommendation list corresponding to the target product information. For newly added products after the previous calculation cycle and before obtaining the recommendation request, the server obtains the relationship graph data of newly added product information and partial product information from the full product information after the previous calculation cycle; when the relationship graph data contains product information from the target recommendation list, it indicates that there are related or similar products of the newly added product in the target recommendation list, and updating the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information. Based on this method, new products can be updated in the user's recommendation list without waiting for the next calculation cycle to generate the list database offline. Compared to adding new product information to the full product information in real time to calculate the global recommendation list, this method simplifies the computation, shortens the calculation time, and solves the problem that users cannot immediately see new products in their recommendation list after they are launched, resulting in low accuracy of recommendation results and poor recommendation performance for new products.

[0083] The following uses Figure 3 The illustrated embodiment provides a detailed explanation of one implementation method for constructing relational graph data in Implementation 1.

[0084] Figure 3 A flowchart illustrating Embodiment 2 of the information push method provided in this application;

[0085] Step S301: Select the associated product information and similar product information corresponding to the newly added product information from the full product information.

[0086] The server obtains the feature information of the newly added product information and calculates the similarity between the feature information of the newly added product information and the feature information of each product information in the full product information.

[0087] Select product information with a similarity greater than a preset threshold from all product information as similar product information;

[0088] Use the database of similar related product information to determine the recommended list corresponding to similar product information;

[0089] The product information in the recommendation list corresponding to similar related product information is used as the related product information.

[0090] Optionally, the server obtains the feature information of the newly added product information and calculates the similarity between the feature information of the newly added product information and the feature information of each product in the full product information, specifically including:

[0091] The feature vectors of each product in the total product information are determined based on the feature information of each product in the total product information, and the feature vectors of newly added product information are determined based on the feature information of newly added product information.

[0092] Optionally, the cosine similarity between the feature vector of the newly added product information and the feature vector of each product information in the full product information can be calculated. If the cosine value of the angle between the two feature vectors is closer to 1, it indicates that the angle between the two feature vectors is closer to 0 degrees, and the two feature vectors are more similar. That is, the similarity between the feature information of the newly added product information and the feature information of the product information in the full product information is higher.

[0093] Optionally, the Pearson similarity between the feature vector of the newly added product information and the feature vector of each product information in the full product information can be calculated. If the Pearson correlation coefficient of the two feature vectors is closer to 1, it indicates that the two feature vectors are more similar, that is, the similarity between the feature information of the newly added product information and the feature information of the product information in the full product information is higher.

[0094] Step S302: Construct a relationship graph data based on the similarity between similar product information and newly added product information and the number of users corresponding to the associated product information.

[0095] After selecting associated product information and similar product information corresponding to the newly added product information from the full product information, a relationship graph data is constructed based on the similarity between the similar product information and the newly added product information and the number of users corresponding to the associated product information.

[0096] Figure 7 A schematic diagram of the information push method provided in this application, such as Figure 7 As shown, relational graph data consists of nodes and edges between nodes. Specifically, the process of constructing relational graph data includes:

[0097] Similar product information, related product information, and newly added product information are used as nodes in the relationship graph data; the similarity between similar product information and newly added product information is used as the edge length between the second node corresponding to similar product information and the first node corresponding to newly added product information; the number of users corresponding to related product information is used as the edge length between the second node corresponding to similar product information and the third node corresponding to related product information.

[0098] Optionally, when the second newly added product is launched and the server receives another recommendation request, since there is relationship graph data in the server, the information of the second newly added product, the information of similar products of the second newly added product, and the information of related products of the second newly added product are added to the relationship graph data to form new relationship graph data.

[0099] The information push method provided in this embodiment selects associated product information and similar product information corresponding to the newly added product information from the full product information. A relationship graph is constructed based on the similarity between the similar product information and the newly added product information, and the number of users corresponding to the associated product information, to prepare for real-time updates of the target recommendation list. This application only constructs relationship graph data based on the newly added product information, similar product information, and associated product information, eliminating the need to construct relationship graph data based on the full product information, thus reducing the computational load and improving computational speed.

[0100] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0101] Figure 4 This is a schematic diagram of the structure of an embodiment of the information push device provided in this application; as shown below. Figure 4 As shown, the information push device 40 includes:

[0102] The acquisition module 41 is used to acquire recommendation requests; wherein, the recommendation request includes target product information;

[0103] Processing module 42 is used to search the list database using the target product information to determine the target recommendation list corresponding to the target product information; wherein, the list database is obtained based on the full product information in the previous calculation cycle;

[0104] The acquisition module 41 is also used to acquire the relationship graph data of newly added product information and partial product information in the full product information after the previous calculation cycle;

[0105] The processing module 42 is also used to update the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and part of the product information in the full product information when the relationship graph data contains product information in the target recommendation list.

[0106] Furthermore, before acquiring the relationship graph data of newly added product information and partial product information in the full product information after the previous calculation cycle, the processing module 42 is also used to select the associated product information and similar product information corresponding to the newly added product information from the full product information;

[0107] Processing module 42 is also used to construct relationship graph data based on the similarity between similar product information and newly added product information and the number of users corresponding to associated product information.

[0108] Furthermore, the processing module 42 is also used to obtain the feature information of the newly added product information and calculate the similarity between the feature information of the newly added product information and the feature information of each product information in the full product information;

[0109] Processing module 42 is also used to select product information with a similarity greater than a preset threshold from the full product information as similar product information;

[0110] Processing module 42 is also used to search the list database using similar related product information to determine the recommendation list corresponding to similar product information;

[0111] Processing module 42 is also used to use the product information in the recommendation list corresponding to similar related product information as related product information.

[0112] Furthermore, the processing module 42 is also used to use similar product information, related product information, and newly added product information as nodes in the relationship graph data;

[0113] The processing module 42 is also used to use the similarity between similar product information and newly added product information as the edge length between the second node corresponding to the similar product information and the first node corresponding to the newly added product information.

[0114] Processing module 42 is also used to use the number of users corresponding to the associated product information as the edge length between the second node corresponding to the similar product information and the third node corresponding to the associated product information.

[0115] Furthermore, the processing module 42 is also used to calculate the correlation between the new product information and the benchmark product information based on the path length and path weight between the fourth node corresponding to the benchmark product information and the first node corresponding to the new product information for each benchmark product information.

[0116] The processing module 42 is also used to determine the position of the new product information in the target recommendation list based on the correlation between each baseline product information and the new product information, and to insert the new product information into the target recommendation list according to the position of the new product information in the target recommendation list;

[0117] Among them, the benchmark product information refers to product information that exists both in the target recommendation list and in the relationship graph data.

[0118] Furthermore, before using the target product information lookup list database to determine the target recommendation list corresponding to the target product information, the processing module 42 is also used to obtain the feature information of all user information and the feature information of all product information when the calculation cycle is reached;

[0119] The processing module 42 is also used to update the list database based on the feature information of all user information and the feature information of all product information.

[0120] Furthermore, the sending module 43 is used to send the updated target recommendation list to the client.

[0121] The information push device provided in this embodiment is used to execute the technical solution in any of the foregoing method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0122] Figure 5 This is a schematic diagram of the structure of an electronic device provided in this application. Figure 5 As shown, the electronic device 50 includes:

[0123] Processor 51, memory 52;

[0124] Memory 52 is used to store executable instructions of processor 51;

[0125] The processor 51 is configured to execute the technical solutions of any of the foregoing method embodiments by executing executable instructions.

[0126] Optionally, the memory 52 can be either standalone or integrated with the processor 51.

[0127] Optionally, when the memory 52 is a device independent of the processor 51, the electronic device 50 may further include:

[0128] A bus is used to connect the aforementioned devices.

[0129] The electronic device is used to execute the technical solutions in any of the foregoing method embodiments. Its implementation principle and technical effect are similar, and will not be described again here.

[0130] This application also provides a readable storage medium storing a computer program thereon, which, when executed by a processor, implements the technical solutions provided in any of the foregoing embodiments.

[0131] This application also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement the technical solutions provided in any of the foregoing method embodiments.

[0132] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0133] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. An information push method, characterized in that, The method is applied to a server, and the method includes: Obtain a recommendation request; wherein the recommendation request includes target product information; The target product information is used to search the list database to determine the target recommendation list corresponding to the target product information; wherein the list database is obtained based on all product information in the previous calculation cycle, and the target recommendation list does not include newly added product information; Select the associated product information and the similar product information corresponding to the newly added product information from all the product information; A relationship graph is constructed based on the similarity between the similar product information and the newly added product information and the number of users corresponding to the associated product information; wherein, the relationship graph data uses the newly added product information, the similar product information, and the associated product information as nodes, the similarity as the edge length between the second node corresponding to the similar product information and the first node corresponding to the newly added product information, and the number of users as the edge length between the second node corresponding to the similar product information and the third node corresponding to the associated product information; Obtain the relationship graph data between the newly added product information added after the previous calculation cycle and a portion of the product information in the total product information; When the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information and a portion of the product information from all the product information.

2. The information push method according to claim 1, characterized in that, Selecting associated product information and similar product information corresponding to the newly added product information from all the product information specifically includes: Obtain the feature information of newly added product information, and calculate and determine the similarity between the feature information of the newly added product information and the feature information of each product information in all product information; Select product information with a similarity greater than a preset threshold from all the product information as the similar product information; The similar product information is used to search the list database to determine the recommended list corresponding to the similar product information; The product information in the recommendation list corresponding to the similar product information is used as the associated product information.

3. The information push method according to claim 1 or 2, characterized in that, When the relationship graph data contains product information from the target recommendation list, the target recommendation list corresponding to the target product information is updated using the relationship graph data of the newly added product information. Specifically, this includes: For each baseline product information, the correlation between the new product information and the baseline product information is calculated based on the path length and path weight between the fourth node corresponding to the baseline product information and the first node corresponding to the new product information. Based on the correlation between each baseline product information and the newly added product information, the position of the newly added product information in the target recommendation list is determined, and the newly added product information is inserted into the target recommendation list according to its position in the target recommendation list; The benchmark product information refers to product information that exists both in the target recommendation list and in the relationship graph data.

4. The information push method according to claim 1 or 2, characterized in that, Before using the target product information lookup list database to determine the target recommendation list corresponding to the target product information, the method further includes: When the calculation cycle is reached, acquire the feature information of all user information and the feature information of all product information; The list database is updated based on the feature information of all user information and the feature information of all product information.

5. The information push method according to claim 1 or 2, characterized in that, The method further includes: Send the updated list of recommended targets to the client.

6. An information push device, characterized in that, The device includes: An acquisition module is used to acquire recommendation requests; wherein, the recommendation request includes target product information; The processing module is used to search the list database using the target product information to determine the target recommendation list corresponding to the target product information; wherein the list database is obtained based on all product information in the previous calculation cycle, and the target recommendation list does not include newly added product information; The acquisition module is also used to acquire relationship graph data of newly added product information added after the previous calculation cycle and some product information in the total product information; The processing module is further configured to update the target recommendation list corresponding to the target product information using the relationship graph data of the newly added product information and a portion of the product information in the total product information when the relationship graph data contains product information in the target recommendation list; The processing module is further configured to select, from all the product information, the associated product information corresponding to the newly added product information and the similar product information corresponding to the newly added product information; The processing module is further configured to construct a relationship graph data based on the similarity between the similar product information and the newly added product information and the number of users corresponding to the associated product information; wherein the relationship graph data uses the newly added product information, the similar product information, and the associated product information as nodes, uses the similarity as the edge length between the second node corresponding to the similar product information and the first node corresponding to the newly added product information, and uses the number of users as the edge length between the second node corresponding to the similar product information and the third node corresponding to the associated product information.

7. An electronic device, characterized in that, include: Processor, memory; The memory is used to store the executable instructions of the processor; The processor is configured to execute the information push method according to any one of claims 1 to 5 by executing the executable instructions.

8. 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 information push method according to any one of claims 1 to 5.

Images