Advertisement pushing method and device, pushing server and storage medium

Abstract

The invention relates to an advertisement pushing method and device, a pushing server and a storage medium. The method comprises the following steps: acquiring real-time behavior data of different users on a real-time display advertisement from a display device, taking the different users and the real-time display advertisement as points, determining a connection line between points corresponding to the different users and the point corresponding to the real-time display advertisement and the association degree of two points on the connection line based on the real-time behavior data, generating graph structure data, and displaying the graph structure data on the display device. On the basis of a connecting line on the graph structure data and the association degree of two points on the connecting line, obtaining interested advertisements of a target user, similar interested users of the target user and interested advertisements of the similar interested users from the graph structure data; and obtaining a target push advertisement of the target user from the interested advertisement of the target user and the similar interested users of the target user, and pushing the target push advertisement. Therefore, the timeliness of advertisement pushing is improved, and the accuracy of advertisement pushing is improved by mining the potential interest points of the user.

Description

Technical Field

[0001] This disclosure relates to the field of advertising push technology, and in particular to an advertising push method, apparatus, push server and storage medium. Background Technology

[0002] With the rapid development of advertising technology, users are exposed to a large number of ads every day. How to accurately push ads to users that match their interests and needs poses a significant challenge to those pushing ads.

[0003] In related technologies, pushers generate ad tags by combining various elements of an ad (e.g., ad title, ad copy, ad image, etc.), and then push relevant ads to different user groups based on these tags. For example, if a user likes ads related to "outdoor" or "dogs," the pusher will push ads to that user containing tags such as "outdoor" and "dogs." However, this ad-pushing method has poor accuracy and timeliness, and needs improvement. Summary of the Invention

[0004] To address the aforementioned technical problems, this disclosure provides an advertising push method, apparatus, push server, and storage medium.

[0005] In a first aspect, this disclosure provides an advertising push method applied to a push server, the method comprising: Obtain real-time behavioral data of different users on real-time displayed advertisements from display devices; Using the different users and the real-time displayed advertisement as points, based on the real-time behavior data, the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisement, as well as the degree of association between the two points on the connection, are determined to generate graph structure data. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, the target user's interested advertisements, similar interested users of the target user, and similar interested users' interested advertisements are obtained from the graph structure data. Among them, the points corresponding to the similar interested users and the points corresponding to the target user are connected to the points corresponding to the same real-time display advertisements. The ads that users with similar interests are interested in and the ads that the target users are interested in are filtered to determine the target ads for the target users; The targeted advertisement is pushed to the display device of the target user.

[0006] Secondly, this disclosure provides an advertising push device configured on a push server, the device comprising: The first acquisition module is used to acquire real-time behavioral data of different users on real-time displayed advertisements from the display device; The graph structure generation module is used to determine the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisements based on the real-time behavior data, as well as the degree of association between the two points on the connection, and generate graph structure data. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. The second acquisition module is used to acquire, based on the connection lines on the graph structure data and the degree of correlation between two points on the connection lines, the target user's interested advertisements, similar interested users of the target user, and similar interested users' interested advertisements from the graph structure data, wherein the point corresponding to the similar interested user and the point corresponding to the target user are connected to the point corresponding to the same real-time display advertisement. The push ad determination module is used to filter the ads of interest to users with similar interests and the ads of interest to the target users, and determine the target push ads for the target users; The advertising push module is used to push the target advertisement to the display device of the target user.

[0007] Thirdly, embodiments of this disclosure also provide a push server, which includes: One or more processors; Storage device for storing one or more programs. When one or more programs are executed by one or more processors, the one or more processors implement the methods provided in the first aspect.

[0008] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the method provided in the first aspect.

[0009] The technical solution provided in this disclosure has the following advantages compared with the prior art: This disclosure discloses an advertising push method, apparatus, push server, and storage medium. It acquires real-time behavioral data of different users' responses to real-time displayed advertisements from a display device. Using different users and real-time displayed advertisements as points, it determines the connections between points corresponding to different users and points corresponding to real-time displayed advertisements based on the real-time behavioral data, as well as the correlation between two points on the connections, generating graph structure data. Since the connections in the graph structure data represent that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the correlation between two points on the same connection represents the corresponding user's level of interest in the real-time displayed advertisement, it acquires, based on the connections and the correlation between two points on the connections, the target user's advertisement of interest, similar users of interest to the target user, and advertisements of interest to similar users of interest from the graph structure data. Since the points corresponding to similar users of interest are connected to the points corresponding to the target user and the same real-time displayed advertisement, by filtering the advertisements of interest to similar users of interest and the advertisements of interest to the target user, it achieves the determination of the target user's target push advertisement and pushes it to the target user's display device. Therefore, it is possible to perceive the behavioral data of different users towards real-time displayed advertisements in real time, analyze the advertisements that a user is interested in, and push advertisements accordingly, thereby improving the timeliness of advertisement push. Furthermore, by analyzing the advertisements that a user is interested in and the advertisements that users with similar interests are interested in within the graph structure data, it is possible to uncover the user's potential interests and improve the accuracy of advertisement push. Attached Figure Description

[0010] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0011] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart illustrating an advertising push method provided in an embodiment of this disclosure; Figure 2 This is a schematic diagram of a graph-structured data provided in an embodiment of the present disclosure; Figure 3 A flowchart illustrating another advertising push method provided in this embodiment of the disclosure; Figure 4 A flowchart illustrating yet another advertising push method provided in this embodiment of the disclosure; Figure 5 A flowchart illustrating another advertising push method provided in an embodiment of this disclosure; Figure 6 A logical schematic diagram of an advertising push method provided in an embodiment of this disclosure; Figure 7 This is a schematic diagram of the structure of an advertising push device provided in an embodiment of the present disclosure; Figure 8 This is a schematic diagram of the structure of a push server provided in an embodiment of this disclosure. Detailed Implementation

[0013] To better understand the above-mentioned objectives, features, and advantages of this disclosure, the solutions disclosed herein will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0014] Numerous specific details are set forth in the following description in order to provide a full understanding of this disclosure, but this disclosure may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some, and not all, of the embodiments of this disclosure.

[0015] In related technologies, methods for pushing ads to different user groups based on tags can specifically involve using diffusion models or large language models to generate ad tags and then pushing ads with appropriate tags to users. However, this method often requires manual maintenance of these tags, cannot discover users' potential interests, and pushes ads to users only after determining the ad tags, resulting in poor real-time performance. Therefore, this ad-pushing method has poor accuracy and timeliness. Furthermore, it does not rely on manual maintenance.

[0016] Some related technologies wait until the next day or even longer after a user views an ad on a display device before pushing that ad to that user. For example, a user might click on a car ad today, but the push server won't push that ad to the user until the next day. Clearly, this method fails to capture the instantaneous changes in user interests, resulting in poor timeliness of ad delivery.

[0017] In related technologies, there is a frequent problem of repeatedly showing the same advertisement to the same user multiple times. This method of ad delivery can easily lead to user fatigue. For example, if a display device shows a user the same advertisement three times, the user may be interested the first time, indifferent the second time, and annoyed the third time. Obviously, this method of repeatedly pushing the same advertisement causes user fatigue and brings a poor viewing experience.

[0018] In related technologies, a display device may show the same advertisement through different applications, only the purpose of displaying the advertisement differs across applications. However, these applications are disconnected and cannot communicate with each other, making it impossible to fully understand which advertisements users are truly interested in. For example, a user might view a phone in application A, search for reviews in application B, and watch an unboxing video in application C. These applications are disconnected and cannot understand whether the user is genuinely interested in the phone.

[0019] To solve the above problems, the following will be combined with... Figures 1-6 The advertising push method provided in this disclosure embodiment will be described. In this disclosure embodiment, the advertising push method can be executed by a push server. The push server can be understood as the service platform of the pusher. Optionally, the push server can be a cloud server or a server cluster.

[0020] Figure 1 A flowchart illustrating an advertising push method provided in an embodiment of this disclosure is shown.

[0021] like Figure 1 As shown, the ad delivery method may include the following steps.

[0022] S110. Obtain real-time behavioral data of different users on real-time displayed advertisements from the display device.

[0023] In this embodiment, during the process of displaying advertisements on the display device, the display device collects real-time behavioral data of users in response to the real-time displayed advertisements and sends the real-time behavioral data to the push server. This allows the push server to analyze the advertisements that users are interested in based on the real-time behavioral data in order to further push advertisements.

[0024] The display device refers to the user's electronic device. Optionally, the display device can be any of the following: mobile phone, tablet, computer, or smartwatch.

[0025] Optionally, the display device can display advertisements pushed by the push server through a variety of different applications, and collect real-time behavioral data of users on real-time displayed advertisements from different applications, and then send the real-time behavioral data to the push server.

[0026] For example, if a user views vehicle price and image ads through app A on their phone, searches for vehicle test drive ads through app B, and searches for vehicle after-sales ads through app C, the display device can obtain user viewing and searching behavior data for vehicle ads from apps A, B, and C as real-time behavioral data.

[0027] In this way, the display device can obtain user behavior data on the same advertisement from different applications and send it to the push server as real-time behavior data. This solves the problem of interoperability between different applications, makes it easier for the push server to obtain more complete data on the advertisement, and thus helps to more comprehensively discover the advertisements that users are interested in.

[0028] Real-time display advertising refers to advertising that is pushed by the pusher to a display device through a push server and then displayed to the user through the display device. Optionally, the specific form of real-time display advertising includes, but is not limited to, images, videos, or text.

[0029] The display device can acquire behavioral data of different users interacting with real-time displayed advertisements through data tracking, which serves as real-time behavioral data. Optionally, real-time behavioral data includes one or more combinations of the following: user browsing time, user click data, user swipe data, user order data, user purchase amount, and user purchase quantity.

[0030] S120. Using different users and real-time displayed advertisements as points, based on real-time behavioral data, determine the connections between the points corresponding to different users and the points corresponding to real-time displayed advertisements, as well as the degree of correlation between the two points on the connection, and generate graph structure data.

[0031] In this embodiment, in order to more accurately identify the ads that each user is interested in, the push server constructs a dynamic association graph based on different users, real-time displayed ads, and real-time behavioral data, so that all users and ads are integrated into a huge relationship network, thereby obtaining graph structure data.

[0032] Graph-structured data is a knowledge representation model that integrates users and advertisements. Connections indicate that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement.

[0033] Optionally, the degree of association can be specifically represented by features such as the width of the connection and the color of the connection.

[0034] As an example, when using width to represent the degree of association between two points on the same line, the higher the degree of interest, the larger the width value, and vice versa.

[0035] For example, when using color to represent the degree of association between two points on the same line, the higher the degree of interest, the darker the color, and vice versa.

[0036] See Figure 2The diagram shown illustrates the structure of graph data. The points in the graph data include user A, ad 1, user B, ad 2, and user C; the lines connecting user A and ad 1; ad 1 and user B; user B and ad 2; and ad 2 and user C. The degree of connection between these lines is represented by their width. It can be seen that... Figure 2 If the width of the line connecting user A and ad 1 is greater than that of other lines, and the widths of the lines connecting ad 1 and user B, user B and ad 2, and ad 2 and user C are equal, then the degree of association between user A and ad 1 is the highest, and the degree of association between user B and ad 1, user B and ad 2, and user C and ad 2 are equal.

[0037] Optionally, the push server can obtain influencing factors that have a substantial effect on calculating the degree of interest from real-time behavioral data, determine the user's degree of interest in the advertisement based on the influencing factors, abstract the user and the advertisement as points respectively, connect the points corresponding to the user and the points corresponding to the advertisement based on the degree of interest, and determine the degree of correlation between the lines connecting the two points.

[0038] Optionally, the aforementioned influencing factors include, but are not limited to, browsing duration, number of views, number of clicks, and purchase amount.

[0039] S130. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, obtain the ads that the target user is interested in, the users with similar interests to the target user, and the ads that users with similar interests are interested in from the graph structure data.

[0040] In this embodiment, since the graph structure data integrates all users and advertisements, as well as the degree of user interest in the advertisements, the push server can mine the advertisements that the target user is interested in from different users, as well as the advertisements that users with similar interests to the target user are interested in, which is beneficial for discovering users' potential interests.

[0041] Among them, the points corresponding to users with similar interests and the points corresponding to the target users are connected to the points corresponding to the same real-time displayed advertisement.

[0042] S140. Filter the ads that users with similar interests are interested in and the ads that the target user is interested in, determine the target push ads for the target user, and push the target push ads to the target user's display device.

[0043] In this embodiment, the push server makes push decisions by analyzing ads that are of interest to users with similar interests and ads that are of interest to target users. Specifically, it makes push targets (i.e., who to push to), push content (i.e., what to push), and push time and place (i.e. push time and device) to push ads that users are interested in for them to view.

[0044] Understandably, since the acquisition of real-time behavioral data, the construction of graph structure data, and the analysis of advertising push are all carried out in real time, the push server can perform everything from perception to decision-making in real time, completing the process in seconds or even milliseconds, which helps improve the timeliness of advertising push.

[0045] Since targeted push ads are ads that interest the target users, when a display device shows a targeted push ad, the target user is highly likely to click on or purchase the product in the targeted push ad. This increases user click-through rate and conversion rate, and also brings better feedback to the pusher. In addition, the ad push method provided in this embodiment is automatically decided by the push server, requiring no manual maintenance, thus reducing labor costs.

[0046] An advertising push method according to an embodiment of this disclosure obtains real-time behavioral data of different users towards real-time displayed advertisements from a display device; using different users and real-time displayed advertisements as points, it determines the connections between the points corresponding to different users and the points corresponding to real-time displayed advertisements based on the real-time behavioral data, and the association between two points on the connections to form a graph, generating graph structure data; since the connections on the graph structure data represent that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement, based on the connections and the degree of association between two points on the connections, it obtains the advertisements of interest to the target user, similar users of the target user, and advertisements of interest to similar users from the graph structure data; since the points corresponding to similar users of interest are connected to the points corresponding to the target user by the points corresponding to the same real-time displayed advertisement, by filtering the advertisements of interest to similar users of interest and the advertisements of interest to the target user, it is possible to determine the target user's target push advertisement and push it to the target user's display device. Therefore, it is possible to perceive the behavioral data of different users towards real-time displayed advertisements in real time, analyze the advertisements that a user is interested in, and push advertisements accordingly, thereby improving the timeliness of advertisement push. Furthermore, by analyzing the advertisements that a user is interested in and the advertisements that users with similar interests are interested in within the graph structure data, it is possible to uncover the user's potential interests and improve the accuracy of advertisement push.

[0047] In some embodiments, after performing S140, the method further includes: when the target user's display device displays the target push advertisement, obtaining user operation data from the target user's display device, wherein the user operation data is determined based on the target user's interaction with the target push advertisement; if the user operation data is data of interest, and the graph structure data does not contain a connection between the point corresponding to the target user and the point corresponding to the target push advertisement, then adding a connection between the point corresponding to the target user and the point corresponding to the target push advertisement in the graph structure data, and determining the degree of association of the added connection according to the user operation data to obtain new graph structure data; if the user operation data is data of non-interest, and the graph structure data contains a connection between the point corresponding to the target user and the point corresponding to the target push advertisement, then deleting the connection between the point corresponding to the target user and the point corresponding to the target push advertisement.

[0048] Interactive actions can include browsing, clicking, placing orders, purchasing, and swiping away. Correspondingly, user action data includes information such as the user's browsing time, click count, order duration, purchase amount, and swipe-away time when viewing the targeted advertisement.

[0049] To determine the reliability of the above-mentioned ad push method, the push server also provides a monitoring mechanism and optimizes the ad push method based on the monitoring results. Specifically, when a targeted push ad is displayed on a target user's display device, the device listens for user interaction, generates user action data, and sends this data to the push server. The push server then determines the type of the user action data. If the user action data is of interest, it means the target user is interested in the targeted push ad, indicating the push ad method is correct. If the graph structure data does not contain connections between the target user's corresponding point and the target push ad's corresponding point, new connections can be added. Specifically, connections are added between the target user's corresponding point and the target push ad's corresponding point, and the correlation of the newly added connections is determined based on the user action data, thereby generating new graph structure data to strengthen the push ad method. If the user action data is of non-interest, it means the target user is not interested in the targeted push ad, indicating the push ad method is inaccurate. If the graph structure data already contains connections between the target user's corresponding point and the target push ad's corresponding point, these connections are deleted, and the targeted push ad will no longer be pushed to the target user's display device.

[0050] In this way, the closed-loop optimization of the ad push method is achieved through this monitoring mechanism, making the ad push method more and more accurate and enabling the pushed ads to automatically adapt to user preferences.

[0051] In another embodiment of this application, the implementation method of S120 will be explained in detail.

[0052] Figure 3 A flowchart illustrating another advertising push method provided in an embodiment of this disclosure is shown.

[0053] like Figure 3 As shown, the ad delivery method may include the following steps.

[0054] S310. Obtain real-time behavioral data of different users on real-time displayed advertisements from the display device.

[0055] S310 is similar to S110, so it will not be described in detail here.

[0056] S320. Based on real-time behavioral data, conduct behavioral analysis on different users who interact with real-time displayed advertisements to determine the different levels of interest of different users in real-time displayed advertisements.

[0057] Since real-time behavioral data is raw data, it is necessary to extract influencing factors from the real-time behavioral data that have a substantial effect on calculating the degree of interest, and then calculate the degree of interest of different users in real-time displayed advertisements based on the influencing factors.

[0058] Optionally, the specific implementation method of S320 includes, but is not limited to, the following methods: obtaining interaction behavior data of different users to real-time displayed advertisements from real-time behavior data; performing interest analysis based on the interaction behavior data and the preset weights corresponding to the interaction behavior data to obtain the degree of interest of different users to real-time displayed advertisements.

[0059] Interactive behavior data can be understood as an influencing factor that has a substantial impact on the degree of interest.

[0060] Optionally, the interaction behavior data includes one or more of the following combinations: browsing duration, number of views, number of clicks, order data, purchase data, etc.

[0061] Optionally, preset weights can be pre-configured for interaction behavior data based on experience.

[0062] Specifically, the push server obtains different interaction behavior data from a large amount of real-time user behavior data, multiplies each interaction behavior data by its corresponding preset weight, and then sums them up to determine the degree of interest of different users in real-time displayed advertisements.

[0063] In one example, a shorter viewing time for a live ad results in a lower weight; in another example, a higher number of clicks on a live ad results in a higher weight; and in yet another example, a shorter time between a user placing an order based on a live ad results in a higher weight.

[0064] In this way, by combining various real-time behavioral data and their corresponding weights, we can quickly analyze the level of interest of different users in real-time displayed advertisements, and at the same time, improve the reliability of interest analysis.

[0065] In some cases, push servers can also analyze the interest characteristics of different users based on different interaction behavior data, and use these interest characteristics to help determine the degree of interest of different users in real-time displayed advertisements.

[0066] Specifically, the push server performs interest analysis based on interaction behavior data and the preset weights corresponding to the interaction behavior data to determine the initial level of interest of different users in real-time displayed ads. Then, it uses the interest characteristics of different users to filter the initial level of interest in order to determine the final level of interest of different users in real-time displayed ads.

[0067] Optionally, interest characteristics include, but are not limited to, theme characteristics, style characteristics, and copywriting characteristics.

[0068] In this way, combining users' interest characteristics helps determine the degree of interest of different users in advertisements, improving the accuracy of the calculation.

[0069] S330: Obtain candidate interest levels greater than a preset threshold from multiple interest levels, and obtain candidate users corresponding to the candidate interest levels from different users.

[0070] In this embodiment, the push server compares multiple levels of interest with preset thresholds. If the level of interest is greater than the preset threshold, it means that the user is interested in the real-time displayed advertisement. The candidate level of interest is then used as the candidate level of interest, and candidate users corresponding to the candidate level of interest are obtained from different users, so as to filter out some users who are sufficiently interested in the advertisement from different users.

[0071] The preset threshold can be a critical condition for determining whether to construct a connection. Optionally, the preset threshold can be a positive number greater than or equal to 0.

[0072] S340. Establish a connection between the point corresponding to the candidate user and the point corresponding to the real-time displayed advertisement, and abstract the candidate's interest level as the degree of correlation between the two points on the connection line to obtain graph structure data.

[0073] In this embodiment, after the push server determines the candidate users, real-time displayed advertisements, and candidate interest levels, it uses a graph structure construction method to establish a connection between the points corresponding to the candidate users and the points corresponding to the real-time displayed advertisements, and abstracts the candidate interest level as the degree of association between the two points on the connection line to generate graph structure data.

[0074] In this way, by mining interaction behavior data from the actual behavior data of different users, and identifying candidate users who are sufficiently interested in real-time displayed ads, it is possible to determine whether to establish a connection between the user's corresponding point and the real-time displayed ad's corresponding point, and to determine the degree of correlation of the connection. This allows multiple users and multiple real-time displayed ads to be integrated into the same graph structure data, facilitating a more comprehensive mining of ads that users are interested in based on the graph structure data.

[0075] S350. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, obtain the ads that the target user is interested in, the users with similar interests to the target user, and the ads that users with similar interests are interested in from the graph structure data.

[0076] Among them, the points corresponding to users with similar interests and the points corresponding to the target users are connected to the points corresponding to the same real-time displayed advertisement.

[0077] S360 filters ads that users with similar interests are interested in and ads that target users are interested in, determines the target push ads for target users, and pushes the target push ads to the target users' display devices.

[0078] S360 is similar to S140, so it will not be described in detail here.

[0079] In another embodiment of this application, the implementation method of S130 will be explained in detail.

[0080] Figure 4 A flowchart illustrating another advertising push method provided in an embodiment of this disclosure is shown.

[0081] like Figure 4 As shown, the ad delivery method may include the following steps.

[0082] S410: Obtain real-time behavioral data of different users on real-time displayed advertisements from the display device.

[0083] S410 is similar to S110, so it will not be described in detail here.

[0084] S420. Using different users and the real-time displayed advertisement as points, based on real-time behavior data, determine the connection between the points corresponding to different users and the points corresponding to the real-time displayed advertisement, as well as the degree of correlation between the two points on the connection, and generate graph structure data.

[0085] The connection indicates that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of correlation between two points on the same connection indicates the degree of interest of the corresponding user in the real-time displayed advertisement.

[0086] S430. Convert the connections on the graph structure data into vectors, and convert the degree of association between two points on the connection into the length of the vector.

[0087] Understandably, when graph structure data cannot be directly read and analyzed by the push server, the push server can first encode the graph structure data, so that the connections in the graph structure data are converted into vectors, and the degree of association between two points on the connection is converted into the length of the vector.

[0088] Specifically, the higher the degree of correlation between two points on a connecting line, the shorter the length of the vector; conversely, the lower the degree of correlation between two points on a connecting line, the longer the length of the vector.

[0089] S440. Based on the first vector containing the endpoint corresponding to the target user and the length of the first vector, determine the advertisement that the target user is interested in, wherein one endpoint of the first vector corresponds to the target user, the other endpoint of the first vector corresponds to the advertisement that the target user is interested in, and the other endpoint of the first vector forms a second vector with the endpoints corresponding to users with similar interests of the target user.

[0090] In this embodiment, for the multiple first vectors where the endpoint corresponding to the target user is located, the push server reads the first vector where the endpoint corresponding to the target user is located and the length of the first vector to obtain the advertisement of interest to the target user.

[0091] Furthermore, the push server obtains a second vector containing ads that interest the target user from multiple vectors, and determines the other endpoint of the second vector as the endpoint corresponding to users with similar interests to the target user.

[0092] S450, Obtain users with similar interests to the target user from the endpoints contained in the second vector.

[0093] In this embodiment, the push server reads the other endpoint of the second vector and uses the other endpoint of the second vector as the endpoint corresponding to users with similar interests to the target user.

[0094] S460. Based on the third vector containing the endpoints of users with similar interests of the target user and the length of the third vector, determine the ads that users with similar interests are interested in. Here, one endpoint of the third vector corresponds to a user with similar interests of the target user, and the other endpoint of the third vector corresponds to an ad that users with similar interests are interested in.

[0095] In this embodiment, for multiple third vectors containing points corresponding to users with similar interests of the target user, the push server reads the third vector containing the endpoint corresponding to the target user and the length of the third vector to obtain the ads that users with similar interests are interested in.

[0096] In this way, graph-structured data is encoded into vector form, and by analyzing the length of the vectors, the ads that a particular user is interested in and the ads that users with similar interests are interested in can be identified.

[0097] S470. Filter the ads that users with similar interests are interested in and the ads that the target user is interested in, determine the target push ads for the target user, and push the target push ads to the target user's display device.

[0098] S470 is similar to S140, so it will not be described in detail here.

[0099] In another embodiment of this application, the implementation method of S140 will be explained in detail.

[0100] Figure 5 A flowchart illustrating another advertising push method provided in an embodiment of this disclosure is shown.

[0101] like Figure 5 As shown, the ad delivery method may include the following steps.

[0102] S510: Obtain real-time behavioral data of different users on real-time displayed advertisements from the display device.

[0103] S510 is similar to S110, so it will not be described in detail here.

[0104] S520. Using different users and the real-time displayed advertisement as points, based on real-time behavior data, determine the connection between the points corresponding to different users and the points corresponding to the real-time displayed advertisement, as well as the degree of correlation between the two points on the connection, and generate graph structure data.

[0105] The connection indicates that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of correlation between two points on the same connection indicates the degree of interest of the corresponding user in the real-time displayed advertisement.

[0106] S530. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, obtain the ads that the target user is interested in, the users with similar interests to the target user, and the ads that users with similar interests are interested in from the graph structure data.

[0107] Among them, the points corresponding to users with similar interests and the points corresponding to the target users are connected to the points corresponding to the same real-time displayed advertisement.

[0108] S540: Extract features from real-time behavioral data to obtain user behavior features.

[0109] In this embodiment, the push server can extract behavioral keywords from real-time behavioral data to determine user behavioral characteristics, or it can use a large language model to extract features from real-time behavioral data to obtain user behavioral characteristics.

[0110] Among them, user behavior characteristics include one or more of the following: click-through rate decay slope, which represents the rate at which user interest declines, and creative fatigue, which represents the number of times a user views the same advertisement.

[0111] For example, if a user clicked on car-related ads 5 times in the last hour but not at all in the last 10 minutes, it indicates that the user's interest in car-related ads has decreased. The rate of decline in user interest in car-related ads can be characterized by the click-through rate decay slope.

[0112] For example, the more times a user sees an ad for the same car category, the greater the user's fatigue with the ad content.

[0113] S550: Utilize user behavior characteristics to filter ads that users with similar interests are interested in and ads that target users are interested in, determine target push ads, and push target push ads to the target user's display device.

[0114] In this embodiment, the push server can filter target push ads from ads of interest to users with similar interests and ads of interest to target users based on click-through rate decay slope and / or creative fatigue.

[0115] By combining the characteristics of interest decay and fatigue, ads that users are truly interested in can be filtered out, while "annoying" irrelevant ads can be reduced. This can proactively protect user experience and maintain long-term advertising efficiency.

[0116] In some embodiments, the push server can also obtain the real-time environmental characteristics of the display device; then, the specific implementation method of S550 includes: filtering the ads of interest to users with similar interests and the ads of interest to target users based on user behavior characteristics to determine candidate recommended ads; and obtaining recommended ads that match the real-time environmental characteristics from the candidate recommended ads as target push ads.

[0117] Real-time environmental features are used to characterize the environmental context of the target user's display device. Real-time environmental features include one or more of the following: time information representing the time of the display device, and location information representing the location of the display device.

[0118] Specifically, the push server can first filter candidate push ads from ads of interest to users with similar interests and ads of interest to target users based on click-through rate decay slope and / or creative fatigue. Then, by combining time information and / or location information, it can filter target push ads from the candidate push ads.

[0119] For example, if afternoon tea vouchers sell particularly well in region A, then by combining location information, ads related to afternoon tea can be filtered and pushed to increase ad click-through rates and conversion rates.

[0120] By combining interest decay, fatigue characteristics, time information, and location information, dynamic creative and anti-fatigue mechanisms are further optimized, making the final recommended ads more suitable for users and improving the value and entertainment value of the ads.

[0121] In another embodiment of this application, the overall logic of the advertising push method is explained in detail.

[0122] Figure 6 A logical schematic diagram of an advertising push method provided by an embodiment of this disclosure is shown.

[0123] like Figure 6 As shown, the ad delivery method may include the following steps.

[0124] S610, Real-time data collection.

[0125] In this embodiment, the push server obtains real-time behavioral data of different users on real-time displayed advertisements from the display device. This real-time behavioral data may include user browsing time, user click data, user swipe data, user order data, user purchase amount, user purchase quantity, and other advertisement data.

[0126] S620, Real-time Feature Data Extraction.

[0127] In this embodiment, the real-time data is raw data, and the push server can extract features from the real-time data to obtain real-time feature data.

[0128] Optionally, real-time feature data may include one or more of the following combinations: click-through rate decay slope, creative fatigue.

[0129] In this embodiment, the push server can also obtain the real-time environmental characteristics of the display device.

[0130] Optionally, the real-time environmental features include one or more of the following: time information representing the time of the display device and location information representing the location of the display device.

[0131] S630, Construct graph structure data.

[0132] In this embodiment, the push server can use different users and real-time displayed advertisements as points, and determine the connections and correlations between the points corresponding to different users and the points corresponding to real-time displayed advertisements based on real-time behavior data, thereby generating graph structure data.

[0133] S640: Make real-time decisions and push ads.

[0134] In this embodiment, the push server performs interest analysis on target users based on the connections and the degree of association of the connections in the graph structure data. It obtains the target user's interested advertisements, users with similar interests, and the interested advertisements of users with similar interests among different users. Then, it filters the interested advertisements of users with similar interests and the interested advertisements of the target user to determine the target push advertisement for the target user and pushes the advertisement.

[0135] S650, closed-loop optimization.

[0136] The above methods enable personalized ad delivery that is millisecond-level, highly relevant, and adaptive, thus preventing user fatigue.

[0137] This disclosure also provides an advertising push device for implementing the above-described advertising push method, which is described below in conjunction with... Figure 7 The following explanation is provided. In this embodiment, the advertising push device can be executed by a push server. A push server can be understood as a service platform for the push end. Optionally, the push server can be a cloud server or a server cluster.

[0138] Figure 7 A schematic diagram of the structure of an advertising push device provided in an embodiment of this disclosure is shown.

[0139] like Figure 7 As shown, the advertising push device 700 may include: The first acquisition module 710 is used to acquire real-time behavioral data of different users on real-time displayed advertisements from the display device; The graph structure generation module 720 is used to generate graph structure data by taking the different users and the real-time displayed advertisement as points, and determining the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisement based on the real-time behavior data, as well as the degree of association between the two points on the connection. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. The second acquisition module 730 is used to acquire, based on the connection on the graph structure data and the degree of correlation between two points on the connection, the target user's interested advertisement, the target user's similar interested users, and the similar interested users' interested advertisement from the graph structure data, wherein the point corresponding to the similar interested user and the point corresponding to the target user are connected to the point corresponding to the same real-time display advertisement. The push advertisement determination module 740 is used to filter the ads of interest to users with similar interests and the ads of interest to target users, and determine the target push advertisements for target users; The advertising push module 750 is used to push the target advertisement to the display device of the target user.

[0140] An advertising push device according to an embodiment of this disclosure acquires real-time behavioral data of different users towards real-time displayed advertisements from a display device; using different users and real-time displayed advertisements as points, it determines the connections between the points corresponding to different users and the points corresponding to real-time displayed advertisements based on the real-time behavioral data, and the association between two points on the connections to form a graph, generating graph structure data; since the connections on the graph structure data represent that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement, based on the connections and the degree of association between two points on the connections, it acquires the advertisements of interest to the target user, similar users of the target user, and advertisements of interest to similar users from the graph structure data; since the points corresponding to similar users of interest are connected to the points corresponding to the target user by the points corresponding to the same real-time displayed advertisement, by filtering the advertisements of interest to similar users of interest and the advertisements of interest to the target user, it is possible to determine the target user's target push advertisement and push it to the target user's display device. Therefore, it is possible to perceive the behavioral data of different users towards real-time displayed advertisements in real time, analyze the advertisements that a user is interested in, and push advertisements accordingly, thereby improving the timeliness of advertisement push. Furthermore, by analyzing the advertisements that a user is interested in and the advertisements that users with similar interests are interested in within the graph structure data, it is possible to uncover the user's potential interests and improve the accuracy of advertisement push.

[0141] In some embodiments of this disclosure, the graph structure generation module 720 includes: The first determining unit is used to perform behavioral analysis on different users interacting with the real-time displayed advertisement based on the real-time behavioral data, and to determine the different users' multiple degrees of interest in the real-time displayed advertisement; The first acquisition unit is used to acquire candidate interest levels that are greater than a preset threshold from the plurality of interest levels, and to acquire candidate users corresponding to the candidate interest levels from the different users; The first generation unit is used to establish a connection between the point corresponding to the candidate user and the point corresponding to the real-time displayed advertisement, and to abstract the candidate interest level as the correlation between two points on the connection line to obtain the graph structure data.

[0142] In some embodiments of this disclosure, the first determining unit is specifically used for: From the real-time behavior data, obtain the interaction behavior data of different users with the real-time displayed advertisement; Interest analysis is performed based on the interaction behavior data and the preset weights corresponding to the interaction behavior data to obtain the degree of interest of different users in the real-time displayed advertisement.

[0143] In some embodiments of this disclosure, the second acquisition module 730 is specifically used for: The connections on the graph structure data are converted into vectors, and the degree of association between two points on the connection is converted into the length of the vector; Based on the first vector containing the endpoint corresponding to the target user and the length of the first vector, the ads that the target user is interested in are determined, wherein one endpoint of the first vector corresponds to the target user, the other endpoint of the first vector corresponds to the ads that the target user is interested in, and the other endpoint of the first vector forms a second vector with the endpoints corresponding to users with similar interests of the target user; Obtain users with similar interests to the target user from the endpoints contained in the second vector; Based on the third vector containing the endpoints of users with similar interests to the target user and the length of the third vector, the ads that users with similar interests are interested in are determined, wherein one endpoint of the third vector corresponds to a user with similar interests to the target user, and the other endpoint of the third vector corresponds to an ad that users with similar interests are interested in.

[0144] In some embodiments of this disclosure, the push advertisement determination module 740 includes: The second acquisition unit is used to extract features from the real-time behavior data to obtain user behavior features, wherein the user behavior features include one or more of the following: click-through rate decay slope, which represents the rate of decline in user interest, and material fatigue, which represents the number of times a user views the same advertisement. The second determining unit is used to filter the ads of interest to users with similar interests and the ads of interest to the target user using the user behavior characteristics, and to determine the target push ad.

[0145] In some embodiments of this disclosure, the push advertisement determination module 740 further includes: The third acquisition unit is used to acquire the real-time environmental features of the display device, wherein the real-time environmental features include one or more of the following: time information representing the time of the display device and location information representing the location of the display device. The second determining unit is specifically used for: Based on the user behavior characteristics, ads that users with similar interests are interested in and ads that the target users are interested in are filtered to determine candidate recommended ads; Recommended ads that match the real-time environmental features are obtained from the candidate recommended ads and used as the target push ads.

[0146] In some embodiments of this disclosure, the device further includes: The third acquisition module is used to acquire user operation data from the target user's display device when the target push advertisement is displayed on the target user's display device, wherein the user operation data is determined based on the target user's interaction with the target push advertisement; An addition module is used to add a connection between the point corresponding to the target user and the point corresponding to the target push advertisement in the graph structure data if the user operation data is data of interest and the graph structure data does not contain a connection between the point corresponding to the target user and the point corresponding to the target push advertisement. The module also determines the degree of association of the added connection based on the user operation data to obtain new graph structure data. The deletion module is used to delete the connection between the point corresponding to the target user and the point corresponding to the target push advertisement if the user operation data is non-interesting operation data and the graph structure data contains the connection between the point corresponding to the target user and the point corresponding to the target push advertisement.

[0147] It should be noted that, Figure 7 The advertising push device 700 shown can perform... Figures 1-6 The various steps in the method embodiment shown are implemented. Figures 1-6 The processes and effects in the method embodiments shown are not described in detail here.

[0148] Figure 8 A schematic diagram of the structure of a push server provided in an embodiment of this disclosure is shown.

[0149] like Figure 8 As shown, the push server may include a processor 801 and a memory 802 storing computer program instructions.

[0150] Specifically, the processor 801 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0151] Memory 802 may include a large-capacity storage device for advertising or instructions. For example, and not limitingly, memory 802 may include a hard disk drive (HDD), a floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or a Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 802 may include removable or non-removable (or fixed) media. Where appropriate, memory 802 may be internal or external to the integrated gateway device. In a particular embodiment, memory 802 is a non-volatile solid-state memory. In a particular embodiment, memory 802 includes read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (Electrically Programmable ROM, EPROM), an electrically erasable programmable PROM (EEPROM), an electrically alterable ROM (EAROM), or flash memory, or a combination of two or more of these.

[0152] The processor 801 acquires and executes computer program instructions stored in the memory 802 to perform the steps of the advertising push method provided in the embodiments of this disclosure.

[0153] In one example, the push server may also include a transceiver 803 and a bus 804. Wherein, as... Figure 8 As shown, the processor 801, memory 802 and transceiver 803 are connected via bus 804 and communicate with each other.

[0154] Bus 804 includes hardware, software, or both. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Extended Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), a Hyper Transport (HT) interconnect, an Industrial Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a MicroChannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local Bus (VLB) bus, or other suitable buses, or a combination of two or more of these. Where appropriate, bus 804 may include one or more buses. Although specific buses are described and illustrated in the embodiments of this application, this application considers any suitable bus or interconnection.

[0155] The following are embodiments of a computer-readable storage medium provided in this disclosure. This computer-readable storage medium belongs to the same inventive concept as the advertising push method in the above embodiments. For details not described in detail in the embodiments of the computer-readable storage medium, please refer to the embodiments of the advertising push method described above.

[0156] This embodiment provides a storage medium containing computer-executable instructions. When executed by a computer processor, these instructions are used to perform an advertising push method applied to a push server. The method includes: Obtain real-time behavioral data of different users on real-time displayed advertisements from display devices; Using the different users and the real-time displayed advertisement as points, based on the real-time behavior data, the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisement, as well as the degree of association between the two points on the connection, are determined to generate graph structure data. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, the target user's interested advertisements, similar interested users of the target user, and similar interested users' interested advertisements are obtained from the graph structure data. Among them, the points corresponding to the similar interested users and the points corresponding to the target user are connected to the points corresponding to the same real-time display advertisements. The system filters the ads that users with similar interests are interested in and the ads that the target user is interested in, determines the target push ads for the target user, and pushes the target push ads to the target user's display device.

[0157] Of course, the computer-executable instructions provided in the embodiments of this disclosure are not limited to the above-described method operations, but can also perform related operations in the information push method provided in any embodiment of this disclosure.

[0158] Based on the above description of the implementation methods, those skilled in the art can clearly understand that this disclosure can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this disclosure, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer cloud platform (which may be a personal computer, a server, or a network cloud platform, etc.) to execute the information push method provided in the various embodiments of this disclosure.

[0159] Note that the above description is merely a preferred embodiment and the technical principles employed in this disclosure. Those skilled in the art will understand that this disclosure is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this disclosure. Therefore, although this disclosure has been described in detail through the above embodiments, it is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this disclosure, and the scope of this disclosure is determined by the scope of the appended claims.

Claims

1. An advertising push method, characterized in that, Applied to a push server, the method includes: Obtain real-time behavioral data of different users on real-time displayed advertisements from display devices; Using the different users and the real-time displayed advertisement as points, based on the real-time behavior data, the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisement, as well as the degree of association between the two points on the connection, are determined to generate graph structure data. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. Based on the connections on the graph structure data and the degree of correlation between two points on the connections, the target user's interested advertisements, similar interested users of the target user, and similar interested users' interested advertisements are obtained from the graph structure data. Among them, the points corresponding to the similar interested users and the points corresponding to the target user are connected to the points corresponding to the same real-time display advertisements. The system filters the ads that users with similar interests are interested in and the ads that the target user is interested in, determines the target push ads for the target user, and pushes the target push ads to the target user's display device.

2. The method according to claim 1, characterized in that, The process involves using the different users and the real-time displayed advertisements as points, determining the connections between the points corresponding to the different users and the points corresponding to the real-time displayed advertisements based on the real-time behavior data, and the degree of correlation between the two points on the connections, to generate graph structure data, including: Based on the real-time behavioral data, behavioral analysis is performed on different users who interact with the real-time displayed advertisement to determine the different levels of interest of the different users in the real-time displayed advertisement; From the plurality of interest levels, obtain candidate interest levels that are greater than a preset threshold, and from the different users, obtain candidate users corresponding to the candidate interest levels; A connection is established between the point corresponding to the candidate user and the point corresponding to the real-time displayed advertisement, and the candidate interest level is abstracted as the correlation between the two points on the connection line to obtain the graph structure data.

3. The method according to claim 2, characterized in that, The step of analyzing the behavior of different users interacting with the real-time displayed advertisement based on the real-time behavior data to determine the different users' levels of interest in the real-time displayed advertisement includes: From the real-time behavior data, obtain the interaction behavior data of different users with the real-time displayed advertisement; Interest analysis is performed based on the interaction behavior data and the preset weights corresponding to the interaction behavior data to obtain the degree of interest of different users in the real-time displayed advertisement.

4. The method according to claim 1, characterized in that, The step of obtaining the target user's interest ads, similar interest users, and similar interest ads from different users based on the connections on the graph structure data and the correlation between two points on the connections includes: The connections on the graph structure data are converted into vectors, and the degree of association between two points on the connection is converted into the length of the vector; Based on the first vector containing the endpoint corresponding to the target user and the length of the first vector, the advertisements that the target user is interested in are determined, wherein one endpoint of the first vector corresponds to the target user, the other endpoint of the first vector corresponds to the advertisements that the target user is interested in, and the other endpoint of the first vector forms a second vector with the endpoints corresponding to users with similar interests of the target user; Obtain users with similar interests to the target user from the endpoints contained in the second vector; Based on the third vector containing the endpoints of users with similar interests to the target user and the length of the third vector, the ads that users with similar interests are interested in are determined, wherein one endpoint of the third vector corresponds to a user with similar interests to the target user, and the other endpoint of the third vector corresponds to an ad that users with similar interests are interested in.

5. The method according to claim 1, characterized in that, The step of filtering ads of interest to users with similar interests and ads of interest to target users to determine the target ads for target users includes: Feature extraction is performed on the real-time behavioral data to obtain user behavior features, wherein the user behavior features include one or more of the following: click-through rate decay slope, which represents the rate of decline in user interest, and material fatigue, which represents the number of times a user views the same advertisement. Using the user behavior characteristics, ads that interest users with similar interests and ads that interest target users are filtered to determine the target push ads.

6. The method according to claim 5, characterized in that, Also includes: The real-time environmental characteristics of the display device are obtained, wherein the real-time environmental characteristics include one or more of the following: time information representing the time of the display device and location information representing the location of the display device; The step of using the user behavior characteristics to filter ads of interest to users with similar interests and ads of interest to target users to determine the target push ads includes: Based on the user behavior characteristics, ads that users with similar interests are interested in and ads that the target users are interested in are filtered to determine candidate recommended ads; Recommended ads that match the real-time environmental features are obtained from the candidate recommended ads and used as the target push ads.

7. The method according to claim 1, characterized in that, Also includes: When the target push advertisement is displayed on the target user's display device, user operation data is obtained from the target user's display device, wherein the user operation data is determined based on the target user's interaction with the target push advertisement; If the user operation data is data of interest, and the graph structure data does not contain a connection between the point corresponding to the target user and the point corresponding to the target push advertisement, then a connection between the point corresponding to the target user and the point corresponding to the target push advertisement is added to the graph structure data, and the degree of association of the added connection is determined according to the user operation data to obtain new graph structure data. If the user operation data is non-interesting operation data, and the graph structure data contains the connection between the point corresponding to the target user and the point corresponding to the target push advertisement, then the connection between the point corresponding to the target user and the point corresponding to the target push advertisement is deleted.

8. An advertising push device, characterized in that, Configured on a push server, the device includes: The first acquisition module is used to acquire real-time behavioral data of different users on real-time displayed advertisements from the display device; The graph structure generation module is used to determine the connection between the points corresponding to the different users and the points corresponding to the real-time displayed advertisements based on the real-time behavior data, as well as the degree of association between the two points on the connection, and generate graph structure data. The connection represents that the real-time displayed advertisement is an advertisement of interest to the corresponding user, and the degree of association between the two points on the same connection represents the degree of interest of the corresponding user in the real-time displayed advertisement. The second acquisition module is used to acquire, based on the connection lines on the graph structure data and the degree of correlation between two points on the connection lines, the target user's interested advertisements, similar interested users of the target user, and similar interested users' interested advertisements from the graph structure data, wherein the point corresponding to the similar interested user and the point corresponding to the target user are connected to the point corresponding to the same real-time display advertisement. The push ad determination module is used to filter the ads of interest to users with similar interests and the ads of interest to the target users, and determine the target push ads for the target users; The advertising push module is used to push the target advertisement to the display device of the target user.

9. A push server, characterized in that, include: processor; Memory, used to store executable instructions; The processor is configured to retrieve the executable instructions from the memory and execute the executable instructions to implement the method of any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The storage medium stores a computer program that, when executed by a processor, causes the processor to implement the method described in any one of claims 1-7.

Images