Live broadcast platform information push method and its device, equipment, medium, and product

Abstract

The present application discloses a method for pushing information on a live broadcast platform and its apparatus, equipment, medium, and product. The method includes: inputting advertising materials into an advertising classification model, and storing the advertising materials and the advertising type labels output by the model in an advertising database; inputting the follow-up content into the advertising classification model, and storing the follow-up content and the advertising type labels output by the model in a content database; responding to an advertising push request acting on a target user terminal, obtaining advertising materials that match the characteristics of the target user terminal from the advertising database to generate an advertising message for push; responding to a platform entry event of the target user terminal, obtaining follow-up content that matches the advertising materials responded to by the target user terminal entering the live broadcast platform from the content database and pushing it to the target user terminal. The present application classifies advertisements and content in the platform using the same model, and quickly and accurately recommends platform content that matches the type of advertisements received by users.

Description

Technical Field

[0001] The present application relates to the field of Internet live broadcasting, and in particular to a method for pushing information on a live broadcasting platform. In addition, the present application also relates to corresponding devices, equipment, non-volatile storage media and computer program products of the method. Background Art

[0002] Existing Internet platforms often need to place advertisements to Internet users to increase their own platform's exposure on the Internet, attract users from other Internet platforms, increase the number of users on their own platforms, or recall old users of their own platforms on the Internet.

[0003] However, the advertising business of most platforms only involves advertising delivery and does not involve the business after users enter the platform through advertising. After users enter the platform through the advertisements pushed by the platform, the platform often requires users to select the type of content they are interested in before recommending corresponding content to users for browsing. For example, when a live broadcast platform pushes advertisements of the live broadcast platform to users, users need to select the type of live broadcast they are interested in when entering the live broadcast platform. Only then will the live broadcast platform push the live broadcast room browsing page containing the live broadcast room they are interested in to users. This process takes a lot of time and is relatively cumbersome for users, and the acceptance is not high, causing some users to quit the platform, resulting in the loss of users on the platform and the failure of the advertisements placed by the platform to achieve the expected results.

[0004] In response to the above-mentioned problems, the applicant has made corresponding explorations based on this issue. Summary of the Invention

[0005] The purpose of this application is to meet the needs of the prior art or overcome at least some of the shortcomings of the prior art and to provide a live broadcast platform information push method and its corresponding device, electronic device, non-volatile storage medium and computer program product.

[0006] In order to achieve the purpose of this application, the following technical solutions are adopted:

[0007] A method for pushing information on a live broadcast platform is proposed to meet the purpose of this application, comprising the following steps:

[0008] Inputting a plurality of advertising creatives into an advertisement classification model that has been trained to convergence, obtaining advertisement type labels output by the advertisement classification model for each of the advertising creatives, and storing the advertising creatives and their advertisement type labels in an advertisement database;

[0009] Inputting multiple contracted contents existing in the live broadcast platform into the advertisement classification model, obtaining the advertisement type labels of the respective contracted contents output by the advertisement classification model, and storing the contracted contents and their advertisement type labels in a content database in correspondence with each other;

[0010] In response to an advertisement push request from a target user terminal, obtaining user characteristic information corresponding to the target user terminal, obtaining advertisement materials having advertisement type tags matching the characteristics of the user characteristic information from the advertisement database, and generating an advertisement message containing the advertisement materials and pushing it to the target user terminal;

[0011] In response to the platform entry event acting on the target user terminal, the advertising material in response to the target user terminal entering the live broadcast platform is determined, and according to the advertising type tag corresponding to the advertising material, the follow-up content with the advertising type tag is obtained from the content database and pushed to the target user terminal.

[0012] In a further embodiment, after the step of acquiring the corresponding advertisement type tag from the content database and pushing the corresponding content with the advertisement type tag to the target user terminal based on the advertisement type tag corresponding to the advertisement material, the following post-processing steps are performed by the server:

[0013] Monitoring user behavior data of the target user terminal in the live broadcast platform, and updating comprehensive return ratio information of the advertisement type corresponding to the advertisement type tag of the advertisement material contained in the advertisement message according to the user behavior data;

[0014] According to the latest updated comprehensive return ratio information for the advertisement type, the ranking position of the advertisement type in the advertisement type ranking list is updated. The advertisement type ranking list contains advertisement types that are sorted according to their comprehensive return ratio information.

[0015] In a further embodiment, the step of inputting a plurality of advertising creatives into an advertisement classification model that has been trained to convergence and obtaining advertisement type labels output by the advertisement classification model to which the respective advertising creatives belong includes the following steps performed by the advertisement classification model:

[0016] Pre-formatting the advertising material;

[0017] Calling a feature extraction model that has been trained to convergence, and extracting a feature vector of the advertising creative by the extraction model;

[0018] A label classification model that has been trained to convergence is called to calculate the probability that the feature vector corresponds to each advertisement type label, and the advertisement type label with the highest probability is used as the advertisement type label to which the advertisement material belongs.

[0019] In a further embodiment, the step of obtaining user characteristic information corresponding to the target user terminal, obtaining advertising materials with advertising type tags matching the user characteristic information from the advertising database, and generating an advertising message containing the advertising materials and pushing it to the target user terminal includes the following steps performed by the server:

[0020] Acquire user characteristic information corresponding to the target user terminal from a user database, and determine the user type corresponding to the target user terminal based on the user characteristic information;

[0021] querying the advertisement database for one or more advertisement type tags corresponding to the user type, and obtaining advertisement materials corresponding to each of the advertisement type tags and stored in the advertisement database;

[0022] An advertisement message including each of the advertisement materials is generated and pushed to the target user terminal.

[0023] In a further embodiment, the step of responding to the platform entry event acting on the target user terminal, determining the advertising material in response to the target user terminal entering the live broadcast platform, and, based on the advertising type tag corresponding to the advertising material, acquiring the follow-up content with the advertising type tag from the content database and pushing it to the target user terminal, includes the following steps performed by the server:

[0024] In response to a platform entry event triggered by the target user terminal through any advertising material in the advertising message, obtaining an advertising material identifier corresponding to the platform entry event;

[0025] Determine the advertising material pointed to by the advertising material identifier, and query the advertising type label corresponding to the advertising material stored in the advertising database;

[0026] Acquire one or more follow-up contents with the advertisement type tag from the content database, and push the follow-up contents to the target user terminal.

[0027] In a further embodiment, the step of acquiring, from the content database and pushing the accompanying content having the advertisement type tag corresponding to the advertisement material to the target user terminal, includes the following steps performed by the target user terminal:

[0028] Receiving the inherited content pushed by the server, parsing the live broadcast room characteristic information contained in the inherited content, wherein the live broadcast room characteristic information includes the live broadcast room name, the anchor user name, and the anchor user avatar;

[0029] The live broadcast room characteristic information is output and displayed in a graphical user interface, and in response to a live broadcast room entry instruction executed on the live broadcast room characteristic information, a live broadcast room page corresponding to the live broadcast room characteristic information is loaded.

[0030] A live broadcast platform information push device proposed for the purpose of this application includes:

[0031] An advertising material classification module is configured to input a plurality of advertising materials into an advertising classification model that has been trained to convergence, obtain advertising type labels output by the advertising classification model to which the advertising materials belong, and store the advertising materials and their advertising type labels in an advertising database;

[0032] The acceptance content classification module is used to input multiple acceptance contents existing in the live broadcast platform into the advertisement classification model, obtain the advertisement type labels of the acceptance contents output by the advertisement classification model, and store the acceptance contents and their advertisement type labels in the content database in correspondence;

[0033] An advertising message push module is configured to respond to an advertising push request from a target user terminal, obtain user characteristic information corresponding to the target user terminal, obtain advertising materials with advertising type tags matching the user characteristic information from the advertising database, and generate an advertising message containing the advertising materials and push it to the target user terminal;

[0034] The takeover content push module is used to respond to the platform entry event acting on the target user terminal, determine the advertising material in response to the target user terminal entering the live broadcast platform, and obtain the takeover content with the advertising type tag from the content database according to the advertising type tag corresponding to the advertising material and push it to the target user terminal.

[0035] In a further embodiment, the advertising material classification module includes:

[0036] An advertising material pre-processing submodule, used for formatting and pre-processing the advertising material;

[0037] A feature vector extraction submodule is used to call a feature extraction model that has been trained to convergence, and the extraction model extracts a feature vector of the advertising material;

[0038] The advertisement type label matching submodule is used to call the label classification model that has been trained to convergence, calculate the probability that the feature vector corresponds to each advertisement type label, and use the advertisement type label with the highest probability as the advertisement type label to which the advertisement material belongs.

[0039] In a further embodiment, the advertising message push module includes:

[0040] A user type determination submodule is configured to obtain user characteristic information corresponding to the target user terminal from a user database, and determine the user type corresponding to the target user terminal based on the user characteristic information;

[0041] An advertising material query submodule, configured to query the advertising database for one or more advertising type tags corresponding to the user type, and obtain advertising materials corresponding to each of the advertising type tags and stored in the advertising database;

[0042] The advertisement message generation submodule is used to generate an advertisement message containing each of the advertisement materials and push it to the target user terminal.

[0043] In a further embodiment, the content push module includes:

[0044] an advertisement identification determination submodule, configured to respond to a platform entry event triggered by the target user terminal through any advertisement material in the advertisement message, and obtain an advertisement material identification corresponding to the platform entry event;

[0045] An advertisement label determination submodule, configured to determine the advertisement material pointed to by the advertisement material identifier, and query the advertisement type label corresponding to the advertisement material stored in the advertisement database;

[0046] The follow-up content acquisition submodule is used to acquire one or more follow-up contents with the advertisement type tag from the content database, and push the follow-up contents to the target user terminal.

[0047] In a preferred embodiment, the content push module further includes:

[0048] The receiving content receiving submodule is used to receive the receiving content pushed by the server and parse the live broadcast room feature information contained in the receiving content, wherein the live broadcast room feature information includes the live broadcast room name, the anchor user name and the anchor user avatar;

[0049] The live broadcast room loading submodule is used to output and display the live broadcast room characteristic information in a graphical user interface, and respond to the live broadcast room entry instruction executed on the live broadcast room characteristic information to load the live broadcast room page corresponding to the live broadcast room characteristic information.

[0050] An electronic device proposed to meet the purpose of this application includes a central processing unit and a memory, and the central processing unit is used to call and run a computer program stored in the memory to execute the steps of the live broadcast platform information push method.

[0051] A non-volatile storage medium is proposed to meet the purpose of this application, which stores a computer program implemented according to the live broadcast platform information push method. When the computer program is called and run by a computer, it executes the steps included in its corresponding method.

[0052] In order to solve the above technical problems, an embodiment of the present application also provides a computer program product, including a computer program and computer instructions. When the computer program and computer instructions are executed by a processor, the processor executes the steps of the above-mentioned live broadcast platform information push method.

[0053] Compared with the prior art, the advantages of this application are as follows:

[0054] This application classifies platform advertisements and platform content into the same classification model so that the content on the platform can be used as the follow-up content of advertisements. When a user enters the platform through an advertisement pushed by the platform, the platform will quickly and accurately recommend corresponding platform content to the user for browsing based on the advertisement to which the user responds, so as to improve the browsing experience of the user after entering the platform through the advertisement. The advertisement delivered by the platform to the user matches the user's characteristics. The follow-up content recommended by the platform to the user is not only related to the advertising business, but also consistent with the user's characteristics, without the need for the user to perform complicated interest editing. After entering the platform, the user will immediately receive the content recommended by the platform for browsing. For example, after a user enters the live broadcast platform by touching a live broadcast advertisement of a certain live broadcast type, the live broadcast platform will push the live broadcast room page corresponding to the live broadcast type touched to the user, effectively improving the user experience, and for the platform, it can effectively improve the conversion rate of platform users.

[0055] Secondly, this application determines the advertising type labels of advertising materials and undertaken content through an advertising classification model constructed based on a neural network model, that is, the advertising classification model is applied to both advertising classification business and platform content classification, so that platform content can be used for undertaking advertising business, and because the types of advertisements and platform content are determined by the same model, the platform can quickly and accurately match users with platform content that matches the advertisements they receive and recommend them.

[0056] Additional aspects and advantages of the present application will be given in part in the following description, which will become apparent from the following description, or will be learned through practice of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The above and / or additional aspects and advantages of the present application will become apparent and easily understood from the following description of the embodiments in conjunction with the accompanying drawings, in which:

[0058] Figure 1 A schematic diagram of a typical network deployment architecture for implementing the technical solution of this application;

[0059] Figure 2 This is a flowchart of a typical embodiment of the live platform information push method of the present application;

[0060] Figure 3 A schematic diagram of a graphical user interface for outputting and displaying advertisement messages to a target user terminal of the present application;

[0061] Figure 4 A schematic diagram of a graphical user interface for outputting live broadcast room feature information contained in the received content to the target user terminal of this application;

[0062] Figure 5 Schematic diagram of the graphical user interface of the live broadcast room page in the target user terminal of this application

[0063] Figure 6 A flowchart illustrating the specific implementation steps for updating the ranking of advertising types in this application;

[0064] Figure 7 A schematic diagram of the display style of the window for the advertisement type ranking list in this application;

[0065] Figure 8 A flowchart showing the specific implementation steps for the advertising classification model used in this application to classify advertising materials;

[0066] Figure 9 This is a flowchart of the specific implementation steps for matching corresponding advertising materials from the advertising database and pushing them to the user end;

[0067] Figure 10 This is a flowchart of the specific implementation steps of this application for matching the corresponding advertising materials from the content database and pushing them to the user end;

[0068] Figure 11 A flowchart showing the specific implementation steps for parsing and accepting content on the target user side of this application;

[0069] Figure 12 This is a principle block diagram of a typical embodiment of the live broadcast platform information push device of the present application;

[0070] Figure 13 This is a basic structural block diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0071] The following describes in detail embodiments of the present application, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and are not to be construed as limiting the present application.

[0072] It will be understood by those skilled in the art that, unless expressly stated otherwise, the singular forms "a", "an", "said" and "the" used herein may also include the plural forms. It should be further understood that the term "comprising" used in the specification of the present application refers to the presence of the features, integers, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof. It should be understood that when we refer to an element as being "connected" or "coupled" to another element, it may be directly connected or coupled to the other element, or there may be intermediate elements. In addition, "connected" or "coupled" as used herein may include wireless connections or wireless couplings. The term "and / or" used herein includes all or any units and all combinations of one or more associated listed items.

[0073] It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be understood to have meanings consistent with their meanings in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0074] It will be understood by those skilled in the art that the terms "client," "terminal," and "terminal device" as used herein include both devices that are wireless signal receivers, i.e., devices that only have wireless signal receivers without transmission capabilities, and devices that have receiving and transmitting hardware capable of two-way communication over a two-way communication link. Such devices may include: cellular or other communication devices such as personal computers and tablet computers, which have single-line displays, multi-line displays, or cellular or other communication devices without multi-line displays; PCS (Personal Communications Service), which may combine voice, data processing, fax, and / or data communication capabilities; PDA (Personal Digital Assistant), which may include a radio frequency receiver, a pager, Internet / Intranet access, a web browser, a notepad, a calendar, and / or a GPS (Global Positioning System) receiver; and conventional laptop and / or palmtop computers or other devices, which have and / or include a radio frequency receiver. As used herein, the terms "client," "terminal," or "terminal device" may be portable, transportable, or installed in a vehicle (air, sea, and / or land), or may be adapted and / or configured to operate locally and / or in a distributed manner at any other location on Earth and / or in space. As used herein, the terms "client," "terminal," or "terminal device" may also refer to a communication terminal, an Internet terminal, or a music / video playback terminal, such as a PDA, an MID (Mobile Internet Device), and / or a mobile phone with music / video playback capabilities, or may include a smart TV, a set-top box, or other device.

[0075] The hardware referred to by names such as "server", "client", and "service node" in this application is essentially an electronic device with capabilities equivalent to those of a personal computer. It is a hardware device that has the necessary components revealed by the von Neumann principle, such as a central processing unit (including an arithmetic unit and a controller), a memory, an input device, and an output device. Computer programs are stored in its memory, and the central processing unit loads the program stored in the external memory into the internal memory for execution, executes the instructions in the program, and interacts with the input and output devices to complete specific functions.

[0076] It should be noted that the concept of "server" referred to in this application can also be extended to server clusters. Based on the network deployment principles understood by those skilled in the art, the servers described should be logically divided. In physical space, these servers can be independent of each other but callable through interfaces, or integrated into a single physical computer or a computer cluster. Those skilled in the art should understand this flexibility and should not use it to constrain the implementation of the network deployment method of this application.

[0077] See also Figure 1 , the hardware foundation required for the implementation of the relevant technical solutions of this application can be deployed according to the architecture shown in the figure. The server 80 referred to in this application is deployed in the cloud. As a business server, it can be responsible for further connecting relevant data servers and other servers that provide relevant support, etc., to form a logically related service cluster to provide services for relevant terminal devices such as the smartphone 81 and personal computer 82 shown in the figure or a third-party server (not shown). The smartphone and personal computer can both access the Internet through a well-known network access method and establish a data communication link with the server 80 in the cloud to run terminal applications related to the services provided by the server.

[0078] For the server, the application is usually constructed as a service process, opening the corresponding program interface for remote calls by applications running on various terminal devices. The relevant technical solutions in this application that are suitable for running on the server can be implemented in the server in this way.

[0079] The application mentioned above refers to an application running on a server or terminal device. This application implements the relevant technical solutions of the present application in a programming manner. Its program code can be stored in a non-volatile storage medium that can be recognized by the computer in the form of computer-executable instructions, and can be loaded into the memory by the central processing unit for execution. The relevant device of the present application is constructed by the operation of this application on the computer.

[0080] For the server, the application is usually constructed as a service process, opening the corresponding program interface for remote calls by applications running on various terminal devices. The relevant technical solutions in this application that are suitable for running on the server can be implemented in the server in this way.

[0081] The technical solution in this application suitable for implementation in terminal devices can also be programmed and built into an application that provides live broadcasting as part of its extended functions. The live broadcasting mentioned above refers to a live broadcast room network service implemented based on the aforementioned network deployment architecture.

[0082] The live broadcast room referred to in this application refers to a video chat room implemented by Internet technology, which usually has audio and video broadcast control functions, including anchor users and audience users. Audience users can include registered users who have registered on the platform, or unregistered visitor users; they can be registered users who follow anchor users, or registered or unregistered users who do not follow anchor users. Anchor users and audience users can interact through well-known online interactive methods such as voice, video, and text. Generally, anchor users perform programs for audience users in the form of audio and video streams, and economic transactions can also be generated during the interaction. Of course, the application form of live broadcast rooms is not limited to online entertainment, but can also be extended to other related scenarios, such as education and training scenarios, video conferencing scenarios, product promotion and sales scenarios, and any other scenarios that require similar interactions.

[0083] Those skilled in the art should be aware that although the various methods of this application are described based on the same concept and thus exhibit commonality, unless otherwise specified, these methods can be independently executed. Similarly, the various embodiments disclosed in this application are all based on the same inventive concept. Therefore, concepts with the same expression, as well as concepts that are appropriately transformed for convenience despite different expression, should be understood as equivalent.

[0084] Unless expressly stated to be mutually exclusive, the various embodiments disclosed in this application may be cross-combined with the relevant technical features of the various embodiments to flexibly construct new embodiments, as long as such combination does not deviate from the creative spirit of this application and can meet the needs of the prior art or resolve certain deficiencies in the prior art. Those skilled in the art should be aware of such flexibility.

[0085] See also Figure 2 In a typical embodiment of the present application, a method for pushing information on a live broadcast platform includes the following steps:

[0086] Step S11: Input multiple advertising materials into an advertisement classification model that has been trained to convergence, obtain the advertisement type labels of the advertisement materials output by the advertisement classification model, and store the advertisement materials and their advertisement type labels in an advertisement database.

[0087] The server inputs a plurality of the advertising materials into the advertisement classification model that has been trained to the convergence to obtain the advertisement type labels to which the advertisement outputs output by the advertisement classification model belong, and then stores the advertisement materials and their advertisement type labels as mapping relationship data in the advertisement database.

[0088] The advertising material generally refers to graphic material or video material used for advertising push. The advertising material is generally pushed to the server by the developer's device so that the server can classify the advertising material through the advertising classification model.

[0089] The advertising type label generally refers to a label used to characterize the advertising type of the advertising material. For example, when the advertising material is a picture of a female anchor, the advertising type label determined by the advertising classification model for the advertising material is generally a female anchor advertising type, and the advertising type label generally also includes the feature vector of the corresponding advertising material / subsequent content, so as to improve the feature vector of the advertising material that can be responded to by the target user end in the future, and match the corresponding subsequent content for the target user end to push it.

[0090] The advertising classification model is generally a classification model constructed based on a neural network model, which is used to identify the advertising type corresponding to the advertising material to determine the advertising type label corresponding to the advertising material. Specifically, after the advertising classification model obtains the advertising material pushed by the server, it will pre-process the advertising material. For example, when the advertising material is an advertising image material, the advertising classification model will adjust the advertising material to a size suitable for classification and identification. If the advertising material is an advertising text material, the advertising classification model will remove special characters contained in the advertising text material and perform word segmentation on the advertising text material; when the advertising classification model completes the pre-processing of the advertising material, the advertising classification model will call a feature extraction model that has been trained to convergence, and the extraction model extracts the feature vector of the advertising material. The feature extraction model is generally a convolutional neural network model, such as a CNN or FCN model; after extracting the feature vector of the advertising material, the advertising classification model will call a label classification model that has been trained to convergence, calculate the probability of the feature vector corresponding to each advertising type label, and use the advertising type label with the largest probability as the advertising type label to which the advertising material belongs. The label classification model is generally constructed based on a normalization algorithm, such as a Softmax algorithm.

[0091] After the server obtains the ad type label of the ad creative through the ad classification model, it stores the ad creative and its ad type label in the ad database. The ad database stores multiple mapping relationships between ad creatives and their corresponding ad type labels, allowing the server to subsequently query the ad database to obtain the corresponding ad creative and generate and deliver an ad message based on the ad creative. For details on the generation and delivery of the ad message, please refer to the subsequent steps; this step will not be elaborated on.

[0092] Step S12: Input the multiple undertaking contents existing in the live broadcast platform into the advertisement classification model, obtain the advertisement type labels of the undertaking contents output by the advertisement classification model, and store the undertaking contents and their advertisement type labels in the content database in correspondence with each other.

[0093] The server obtains a plurality of the undertaking contents that have not been classified as advertisements in the live broadcast platform, inputs these undertaking contents into the advertisement classification model to obtain the advertisement type labels corresponding to each of the undertaking contents output by the advertisement classification model, and stores these undertaking contents and their advertisement type labels as mapping relationship data in the content database.

[0094] The described takeover content is the platform content existing in the live broadcast platform, and the described takeover content can be used to take over the advertising messages delivered through the live broadcast platform for users of the platform to provide the corresponding platform content to the user, and the described takeover content corresponds to the business existing in the live broadcast platform. For example, when there is a platform activity business in the live broadcast platform, there will be corresponding takeover content for the platform activity in the live broadcast platform to provide users in the live broadcast platform with corresponding platform activities to participate in; when there is a corresponding live broadcast business in the live broadcast platform, there will be takeover content for the live broadcast room in the live broadcast platform; when there is a streaming media business in the live broadcast platform, there will be corresponding takeover content for streaming media in the live broadcast platform.

[0095] The server collects one or more of the aforementioned undertaking contents that have not been classified as advertisements on the live broadcast platform, and inputs these undertaking contents into the advertisement classification model. The advertisement classification model is the same model as the model used by the server to classify the advertisement materials. The server will pre-process these undertaking contents. For example, when the undertaking content is a live broadcast room, the advertisement classification model will generally identify the advertisement type corresponding to the anchor user picture of the live broadcast room or the live broadcast stream corresponding to the undertaking content. The identification method also calls the feature extraction model and the label classification model to determine the advertisement type label corresponding to the undertaking content.

[0096] After the server obtains the advertisement type label corresponding to the undertaking content output by the advertisement classification model, it stores the undertaking content and the advertisement type label in the content database. The content material library stores a plurality of mapping relationship data consisting of the undertaking content and its corresponding advertisement type labels.

[0097] Step S13: In response to the advertisement push request for the target user terminal, user characteristic information corresponding to the target user terminal is obtained, advertisement materials with advertisement type tags matching the characteristics of the user characteristic information are obtained from the advertisement database, and an advertisement message containing the advertisement materials is generated and pushed to the target user terminal.

[0098] The server responds to the advertising push request acting on the target user terminal, obtains the user feature information corresponding to the target user terminal, obtains one or more advertising materials with advertising type labels similar to the user feature information characteristics from the advertising database, and generates an advertising message containing these advertising materials and pushes it to the target user terminal.

[0099] The advertising push request is generally pushed by the advertiser user end. After the advertiser user end defines the target user end to which the advertisement is to be pushed, the advertising push request pointing to the target user end will be generated and pushed to the server, so that the server can respond to the advertising push request, determine the target user end to which the request points, and then obtain the user feature information of the target user.

[0100] After the server determines the target user terminal to which the advertising push request points, it will obtain the user feature information corresponding to the target user terminal from the user database. The user feature information generally includes gender characteristics, age characteristics, interest characteristics and consumption power characteristics, etc. The user database stores multiple mapping relationship data consisting of user feature information and its corresponding user identifiers. The server constructs the user identifier corresponding to the target user terminal and obtains the user feature information stored corresponding to the user identifier from the user database.

[0101] After the server obtains the user feature information of the target user terminal, it will extract the user features represented by the user feature information to determine the user type corresponding to the target user terminal. For example, when the gender feature in the user feature information is female, the age feature is 24 years old, and the interest feature is sports, the server will extract the user type represented by the user feature information as a young female user type or a sports female user type.

[0102] After the server determines the user type corresponding to the target user terminal based on the user characteristic information, it will obtain advertising materials with advertising type labels that match the user type from the advertising database. For example, when the user type is a live broadcast user type, the server will obtain one or more advertising materials with advertising type labels characterized as live broadcast advertising types from the advertising database. Specifically, the server extracts the user feature vector of the user characteristic information, and extracts the advertising feature vectors corresponding to all the advertising materials in the advertising database, calculates the feature similarity between the user feature vector and each of the advertising feature vectors, and then sorts the advertising feature vectors from large to small according to their feature similarities with the user feature vector, and matches the advertising materials corresponding to the advertising feature vectors that are ranked first or in the top three with the user feature information.

[0103] In addition, the server can also match the corresponding advertising materials for the user feature information through preset matching rules. After the server determines the user type of the target user terminal, it uses preset matching rules, such as matching sports users with sports advertisements or matching food users with food advertisements, to determine one or more advertising type tags corresponding to the user type.

[0104] After the server obtains the advertising materials corresponding to the target user terminal from the advertising database, it generates the advertising message containing the advertising materials and pushes it to the target user terminal, so that after the target user terminal receives the advertising message, it outputs the advertising materials contained in the advertising message to a graphical user interface for display. For example, please refer to Figure 3 , Figure 3 This is a schematic diagram of the target user terminal outputting the received advertisement message to the current graphical user interface for display. Figure 3 The advertising material output in the advertising material display control 301 is the advertising material of the YY Bear Live Room for display.

[0105] Step S14: In response to the platform entry event acting on the target user terminal, determine the advertising material in response to the target user terminal entering the live broadcast platform, and according to the advertising type tag corresponding to the advertising material, obtain the corresponding content with the advertising type tag from the content database and push it to the target user terminal:

[0106] The server responds to the platform entry event acting on the target user terminal, determines the advertising material in response to the target user terminal entering the live broadcast platform, queries the follow-up content with the advertising type tag having the advertising material from the content database, and pushes the follow-up content to the target user terminal.

[0107] The platform entry event is an event triggered when the target user terminal responds to any advertising material contained in the advertising message and enters the live broadcast platform. Specifically, the advertising message will include a plurality of advertising materials output to the graphical user interface of the target user terminal for display. When a user at the target user terminal touches a certain advertising material displayed in the graphical user interface, the advertising material touched by the user is the advertising material responded to by the user entering the live broadcast platform, and the platform entry event includes an advertising material identifier pointing to the advertising material responded to by the target user terminal, so that the server can determine the advertising material responded to by the target user terminal based on the advertising material identifier.

[0108] After the server determines the advertising material responded by the target user terminal, it will query the advertising type label corresponding to the advertising material from the advertising material library, and then query the corresponding content stored corresponding to the advertising type label from the content database, and then push the following content to the target user terminal.

[0109] Please refer to Figure 3 、 Figure 4 and Figure 5 When the advertisement type label corresponding to the advertisement material responded by the target user terminal is a live broadcast advertisement type, if the target user terminal touches Figure 3 The advertising material displayed in the advertising material display control 301 is output as the advertising material of the YY Bear live broadcast room. After entering the live broadcast platform, it will receive the follow-up content containing the live broadcast room feature information pushed by the server. The target user end will parse the live broadcast room feature information contained in the follow-up content, and obtain the live broadcast room name, anchor user name and anchor user avatar and other information contained in the live broadcast room feature information. Its graphical user interface will be as follows Figure 4 As shown, the target user terminal outputs the live broadcast room feature information to the live broadcast room guide window 401 for display, and after a preset time, such as 3 to 4 seconds, the live broadcast room page pointed to by the live broadcast room feature information is loaded for display. Figure 4 The live room browsing page shown in the graphical user interface jumps to Figure 5 The live broadcast room page shown.

[0110] It can be seen from this typical embodiment that this method determines the advertising type labels of advertising materials and the accompanying content through the advertising classification model constructed based on the neural network model, and stores the advertising materials and their advertising type labels in the advertising database in correspondence, and the accompanying content also stores its corresponding advertising type labels in the content database in correspondence, so that the content in the platform can be used as the accompanying content of the advertisement. When the user enters the platform through the advertisement pushed by the platform, the platform will quickly and accurately recommend the corresponding platform content to the user for browsing based on the advertisement responded by the user, so as to improve the browsing experience of the user after entering the platform through the advertisement, and the advertisement delivered by the platform to the user matches the characteristics of the user, then the platform will recommend The follow-up content recommended by users is not only related to the advertising business, but also consistent with the characteristics of the users. Users do not need to perform complicated interest editing. After entering the platform, they immediately receive the content recommended by the platform for browsing, which can effectively improve the conversion rate of platform users. Secondly, this method determines the advertising type labels of the advertising materials and the follow-up content through the advertising classification model, that is, the advertising classification model is applied to the advertising classification business as well as the classification of platform content, so that the platform content can be used for the undertaking of advertising business, and because the types of advertisements and platform content are determined by the same model, the platform can quickly and accurately match users with platform content that matches the advertisements they receive and recommend them.

[0111] The above typical embodiments and their variations fully disclose the implementation scheme of the live platform information push method of the present application. However, various variations of the method can be derived by changing and amplifying some technical means. Other embodiments are briefly described below:

[0112] In one embodiment, please refer to Figure 6 and Figure 7 After the step of acquiring the corresponding advertisement type tag from the content database and pushing the corresponding content with the advertisement type tag to the target user terminal according to the advertisement type tag of the advertisement material, the following post-processing steps are performed by the server:

[0113] Step S15: monitoring the user behavior data of the target user terminal in the live broadcast platform, and updating the comprehensive return ratio information of the advertisement type corresponding to the advertisement type label of the advertisement material contained in the advertisement message according to the user behavior data:

[0114] The server monitors the user behavior data of the target user terminal in the live broadcast platform, and confirms the advertising type label of the advertising material contained in the advertising message pushed to the target user terminal based on the login duration data, platform recharge data and platform consumption data contained in the user behavior data, so as to update the comprehensive return ratio information of the advertising type corresponding to the advertising type label.

[0115] The comprehensive return ratio information refers to the data obtained by converting the user behavior data of the target user terminal. The specific conversion method is generally determined by dividing the numerical value represented by the user behavior data by the numerical value represented by the resources spent by the live broadcast platform on pushing advertisements to the target user terminal. The server recalculates the comprehensive return ratio information based on the latest user behavior data of the target user terminal for update.

[0116] Step S16: Update the ranking position of the advertisement type in the advertisement type ranking list based on the latest updated comprehensive return ratio information for the advertisement type. The advertisement type ranking list contains advertisement types that are sorted according to their comprehensive return ratio information:

[0117] The server updates the ranking position of the advertisement type in the advertisement type ranking list according to the latest updated comprehensive return ratio information of the advertisement type.

[0118] The advertisement type ranking list stores a plurality of mapping relationship data consisting of advertisement types and their corresponding comprehensive return ratio information, and the advertisement type ranking list is sorted in descending order according to the comprehensive return ratio information corresponding to each advertisement type, and the advertisement type ranking list is generally pushed by the server to the advertisement terminal for output display. For details, please refer to Figure 7 , Figure 7 The figure is a schematic diagram showing the display style of a window for outputting and displaying the advertisement type ranking list.

[0119] In this embodiment, the comprehensive return ratio of each advertising type is updated by monitoring the behavior of users who enter the platform through advertisements, and the comprehensive return ratio of each advertising type is visualized in the form of a list, so that the platform can determine the conversion ratio of each advertising type based on the list, determine the advertising type that users like, and then deliver advertisements with higher acceptance to users.

[0120] In one embodiment, please refer to Figure 8 The step of inputting a plurality of advertising creatives into an advertisement classification model that has been trained to convergence and obtaining advertisement type labels output by the advertisement classification model to which the respective advertising creatives belong includes the following steps performed by the advertisement classification model:

[0121] Step S111: formatting and pre-processing the advertising material:

[0122] After the advertising classification model obtains the advertising material pushed by the server, it will perform the formatting preprocessing on the advertising material. For example, when the advertising material is an advertising picture material, the advertising classification model will scale the image size of the advertising picture material to the image size required by the subsequent feature extraction model. When the advertising material is an advertising text material, the advertising classification model will remove the special characters contained in the advertising text material and perform word segmentation processing on the advertising text material.

[0123] Step S112: Calling a feature extraction model that has been trained to convergence, and extracting a feature vector of the advertising material using the extraction model:

[0124] After the advertisement classification model completes the formatting preprocessing of the advertisement material, it calls the feature extraction model that has been trained to convergence. The feature extraction model is generally constructed based on FCN or CNN to extract the feature vector of the advertisement material.

[0125] Step S113: Call the label classification model that has been trained to convergence, calculate the probability that the feature vector corresponds to each advertisement type label, and use the advertisement type label with the highest probability as the advertisement type label to which the advertisement material belongs:

[0126] After the advertising classification model obtains the feature vector of the advertising material, it will call the label classification model that has been converged to calculate the similarity between the feature vector and the characteristics of the advertising type represented by each of the advertising type labels, and then determine the probability that the advertising material corresponds to each of these advertising type labels. The sum of the probabilities that the advertising material corresponds to each of these advertising type labels is generally 1; the label classification model is generally constructed based on a normalization algorithm, such as Softmax.

[0127] After the advertisement classification model obtains the feature vector output by the label classification model and the probability of each advertisement type label, the advertisement type label with the greatest probability is used as the advertisement type label to which the advertisement material corresponding to the feature vector belongs.

[0128] In this embodiment, the advertisement classification model is constructed based on a neural network model to accurately extract advertisement material features and determine the advertisement type to which the advertisement material belongs.

[0129] In one embodiment, please refer to Figure 9 The step of obtaining user characteristic information corresponding to the target user terminal, obtaining advertising materials with advertising type tags matching the user characteristic information from the advertising database, and generating an advertising message containing the advertising materials and pushing it to the target user terminal includes the following steps performed by the server:

[0130] Step S131: Obtain user characteristic information corresponding to the target user terminal from a user database, and determine the user type corresponding to the target user terminal based on the user characteristic information:

[0131] After responding to the advertisement delivery request for the target user terminal, the server will obtain the user characteristic information corresponding to the target user terminal from the user database, and determine the user type corresponding to the target user terminal based on the user characteristic information.

[0132] The user characteristic information generally includes gender characteristics, age characteristics, interest characteristics and consumption power characteristics, etc. The user database stores multiple mapping relationship data consisting of user characteristic information and its corresponding user identification. The server constructs the user identification corresponding to the target user terminal and obtains the user characteristic information stored corresponding to the user identification from the user database.

[0133] After the server obtains the user feature information of the target user terminal, it will extract the user features represented by the user feature information to determine the user type corresponding to the target user terminal. For example, when the gender feature in the user feature information is female, the age feature is 24 years old, and the interest feature is sports, the server will extract the user type represented by the user feature information as a young female user type or a sports female user type.

[0134] Step S132: query the advertisement database for one or more advertisement type tags corresponding to the user type, and obtain the advertisement materials stored in the advertisement database corresponding to each of the advertisement type tags.

[0135] After the server determines the user type of the target user terminal, it will query the advertising material library for one or more advertising type labels corresponding to the user type. Specifically, the server can calculate the similarity probability between the user type and each advertising type through a normalization algorithm, or through preset matching rules, such as matching sports users with sports advertisements or matching food users with food advertisements, to determine one or more advertising type labels corresponding to the user type.

[0136] After determining one or more advertisement type tags corresponding to the user type, the server acquires the advertisement materials stored corresponding to the advertisement type tags from the advertisement material library.

[0137] Step S133: Generate an advertisement message containing each of the advertisement materials and push it to the target user terminal.

[0138] After obtaining one or more advertisement materials corresponding to the user type from the database, the server encapsulates the advertisement materials, generates the advertisement message containing the advertisement materials, and pushes the advertisement message to the target user terminal.

[0139] In this embodiment, the server accurately matches advertising materials for users according to user types, and delivers advertisements that match the characteristics of the users, thereby improving the users' acceptance of advertisements.

[0140] In one embodiment, please refer to Figure 10 The step of responding to the platform entry event acting on the target user terminal, determining the advertising material in response to the target user terminal entering the live broadcast platform, and acquiring, based on the advertising type tag corresponding to the advertising material, the accompanying content with the advertising type tag from the content database and pushing it to the target user terminal includes the following steps executed by the server:

[0141] Step S141: In response to a platform entry event triggered by any advertising material in the advertising message, the target user terminal obtains an advertising material identifier corresponding to the platform entry event.

[0142] When any advertising material in the advertising message pushed by the server enters the live broadcast platform, the target user terminal will trigger the platform entry event containing the advertising material identifier corresponding to the advertising material it responds to, so that the server responds to the platform entry event and obtains the advertising material identifier.

[0143] Step S142: Determine the advertising material pointed to by the advertising material identifier, and query the advertising type label corresponding to the advertising material stored in the advertising database:

[0144] After obtaining the material identifier, the server will determine the advertising material pointed to by the advertising material identifier, and query the advertisement database to obtain the advertisement type label stored corresponding to the advertisement material.

[0145] Step S143: Obtain one or more follow-up contents with the advertisement type tag from the content database, and push the follow-up contents to the target user terminal.

[0146] After obtaining the advertisement type tag from the advertisement database, the server searches for one or more follow-up contents having the advertisement type tag from the content database, and pushes the follow-up contents to the target user terminal.

[0147] In this embodiment, by determining the type of advertisement through which the user enters the platform, platform content matching the advertisement type is pushed to the user, so as to provide the user with content related to the purpose of entering the platform for browsing, thereby improving the user's experience on the platform.

[0148] In one embodiment, please refer to Figures 3 to 5 and Figure 11 The step of acquiring, from the content database, the accompanying content having the advertisement type tag corresponding to the advertisement material and pushing the content to the target user terminal includes the following steps performed by the target user terminal:

[0149] Step S141': receiving the inherited content pushed by the server, parsing the live broadcast room feature information contained in the inherited content, wherein the live broadcast room feature information includes the live broadcast room name, anchor user name, and anchor user avatar:

[0150] When the server determines that the advertisement type responded by the target user terminal entering the live broadcast platform is a live broadcast advertisement type, Figure 3 As shown, the target user terminal passes Figure 3 When the advertising material display control 301 shown in enters the live broadcast platform, the server will push the content of the live broadcast room content to the target user end, and the follow-up content includes the live broadcast room feature information of the live broadcast room corresponding to the advertising material it responds to, so that the target user end can parse the live broadcast room feature information contained in the follow-up content, and output the live broadcast room name, anchor user name and anchor user avatar contained in the live broadcast room feature information to the current live broadcast room for display.

[0151] Step S142': output and display the live broadcast room characteristic information in a graphical user interface, and in response to the live broadcast room entry instruction executed based on the live broadcast room characteristic information, load the live broadcast room page corresponding to the live broadcast room characteristic information:

[0152] After the target user terminal obtains the live broadcast room feature information, Figure 4 As shown, the live broadcast room name, anchor user name and anchor user avatar contained in the live broadcast room characteristic information are output to the live broadcast room guide window 401 for display, and after a preset time, such as 3 to 4 seconds, in response to the live broadcast room entry instruction executed based on the live broadcast room characteristic information, the live broadcast room page pointed to by the live broadcast room link is loaded and displayed according to the live broadcast room link contained in the live broadcast room characteristic information. Figure 4 The live room browsing page shown in the graphical user interface jumps to Figure 5 The live broadcast room page shown.

[0153] In this embodiment, a live broadcast room guide window will first appear in the graphical user interface of the user end to output the live broadcast room characteristics to notify the live broadcast room that is about to be entered, and then the live broadcast room corresponding to the live broadcast room characteristics will be loaded for display, and the advertising live broadcast room that the user is interested in will be automatically loaded for display.

[0154] Furthermore, by functionalizing each step in the method disclosed in the above embodiments, a live broadcast platform information push device of the present application can be constructed. According to this idea, please refer to Figure 12 In a typical embodiment, the device includes: an advertising material classification module 11, which is used to input multiple advertising materials into an advertising classification model that has been trained to convergence, obtain the advertising type labels to which the advertising materials are respectively output by the advertising classification model, and store the advertising materials and their advertising type labels in the advertising database in correspondence; a continuation content classification module 12, which is used to input multiple continuation contents existing in the live broadcast platform into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the continuation contents and their advertising type labels in the content database in correspondence; an advertising consumption module 13, which is used to input multiple continuation contents existing in the live broadcast platform into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the continuation contents and their advertising type labels in the content database in correspondence; an advertising consumption module 14, which is used to input multiple advertising materials into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the advertising consumption modules 14 in the advertising database in correspondence; an advertising consumption module 15, which is used to input multiple advertising materials into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the advertising consumption modules 14 in the advertising database in correspondence; an advertising consumption module 16, which is used to input multiple advertising materials into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the advertising consumption modules 14 in the advertising database in correspondence; an advertising consumption module 17, which is used to input multiple advertising materials into the advertising classification model, obtain the advertising type labels to which the continuation contents are respectively output by the advertising classification model, and store the advertising consumption modules 17 and their advertising type labels in the content database in correspondence; an advertising consumption module 18, which is used to input multiple advertising materials into the advertising classification model, obtain the advertising type labels to which the The information push module 13 is used to respond to the advertising push request acting on the target user terminal, obtain the user characteristic information corresponding to the target user terminal, obtain the advertising material with the advertising type label matching the user characteristic information from the advertising database, and generate an advertising message containing these advertising materials and push it to the target user terminal; the follow-up content push module 14 is used to respond to the platform entry event acting on the target user terminal, determine the advertising material in response to the target user terminal entering the live broadcast platform, and according to the advertising type label corresponding to the advertising material, obtain the follow-up content with the advertising type label from the content database and push it to the target user terminal.

[0155] In one embodiment, the advertising material classification module 11 includes: an advertising material preprocessing submodule, used to format and preprocess the advertising material; a feature vector extraction submodule, used to call a feature extraction model that has been trained to convergence, and the extraction model extracts the feature vector of the advertising material; an advertising type label matching submodule, used to call a label classification model that has been trained to convergence, calculate the probability that the feature vector corresponds to each advertising type label, and use the advertising type label with the highest probability as the advertising type label to which the advertising material belongs.

[0156] In one embodiment, the advertising message push module 13 includes: a user type determination submodule, used to obtain user feature information corresponding to the target user terminal from the user database, and determine the user type corresponding to the target user terminal based on the user feature information; an advertising material query submodule, used to query one or more advertising type tags corresponding to the user type in the advertising database, and obtain the advertising materials corresponding to each of these advertising type tags stored in the advertising database; an advertising message generation submodule, used to generate an advertising message containing each of the advertising materials and push it to the target user terminal.

[0157] In one embodiment, the continuation content push module 14 includes: an advertising identifier determination submodule, which is used to respond to the platform entry event triggered by any advertising material in the advertising message of the target user terminal, and obtain the advertising material identifier corresponding to the platform entry event; an advertising label determination submodule, which is used to determine the advertising material pointed to by the advertising material identifier, and query the advertising type label corresponding to the advertising material stored in the advertising database; a continuation content acquisition submodule, which is used to obtain one or more continuation contents with the advertising type label in the content database, and push these continuation contents to the target user terminal.

[0158] In another embodiment, the takeover content push module 14 also includes: a takeover content receiving sub-module, used to receive the takeover content pushed by the server, and parse out the live broadcast room characteristic information contained in the takeover content, wherein the live broadcast room characteristic information includes the live broadcast room name, anchor user name and anchor user avatar; a live broadcast room loading sub-module, used to output and display the live broadcast room characteristic information in a graphical user interface, and respond to the live broadcast room entry instruction executed by the live broadcast room characteristic information, and load the live broadcast room page corresponding to the live broadcast room characteristic information.

[0159] In order to solve the above technical problems, the embodiment of the present application also provides a computer device for running a computer program implemented according to the live platform information push method. Figure 13 , Figure 13 This is a basic structural block diagram of the computer device in this embodiment.

[0160] like Figure 13As shown, a schematic diagram of the internal structure of a computer device. The computer device includes a processor, a non-volatile storage medium, a memory and a network interface connected via a system bus. Among them, the non-volatile storage medium of the computer device stores an operating system, a database and computer-readable instructions, and the database may store a control information sequence, and when the computer-readable instructions are executed by the processor, the processor may implement a live platform information push method. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor may execute a live platform information push method. The network interface of the computer device is used to connect and communicate with the terminal. Those skilled in the art will understand that Figure 13 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0161] In this embodiment, the processor is used to execute the specific functions of each module / submodule in the live platform information push device of the present invention, and the memory stores the program code and various data required to execute these modules. The network interface is used to transmit data between user terminals or servers. The memory in this embodiment stores the program code and data required to execute all modules / submodules in the live platform information push device, and the server can call the server's program code and data to execute the functions of all submodules.

[0162] The present application also provides a non-volatile storage medium, in which the live broadcast platform information push method is written into a computer program and stored in the storage medium in the form of computer-readable instructions. When the computer-readable instructions are executed by one or more processors, it means that the program is running in the computer, thereby enabling one or more processors to execute the steps of the live broadcast platform information push method of any of the above embodiments.

[0163] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When executed, the program can include the processes in the above-described method embodiments. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0164] In summary, this application uses the same model to classify advertisements and content on the platform, and quickly and accurately recommends platform content that matches the type of advertisements received by users.

[0165] It should be understood that although the steps in the flowcharts of the accompanying drawings are shown in sequence as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in the flowcharts of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a portion of the sub-steps or stages of other steps.

[0166] Those skilled in the art will appreciate that the steps, measures, and schemes in the various operations, methods, and processes discussed in this application may be interchanged, modified, combined, or deleted. Furthermore, other steps, measures, and schemes in the various operations, methods, and processes discussed in this application may also be interchanged, modified, rearranged, decomposed, combined, or deleted. Furthermore, steps, measures, and schemes in the prior art that are similar to those disclosed in this application may also be interchanged, modified, rearranged, decomposed, combined, or deleted.

[0167] The above description is only part of the implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for pushing information on a live broadcast platform, characterized in that: The steps include: Inputting a plurality of advertising creatives into an advertisement classification model that has been trained to convergence, obtaining advertisement type labels output by the advertisement classification model for each of the advertising creatives, and storing the advertising creatives and their advertisement type labels in an advertisement database; Inputting multiple contracted contents existing in the live broadcast platform into the advertisement classification model, obtaining the advertisement type labels of the respective contracted contents output by the advertisement classification model, and storing the contracted contents and their advertisement type labels in a content database in correspondence with each other; In response to an advertisement push request from a target user terminal, obtaining user characteristic information corresponding to the target user terminal, obtaining advertisement materials having advertisement type tags matching the characteristics of the user characteristic information from the advertisement database, and generating an advertisement message containing the advertisement materials and pushing it to the target user terminal; In response to the platform entry event acting on the target user terminal, the advertising material in response to the target user terminal entering the live broadcast platform is determined, and according to the advertising type tag corresponding to the advertising material, the follow-up content with the advertising type tag is obtained from the content database and pushed to the target user terminal, so that the target user terminal executes: receiving the follow-up content pushed by the server, parsing the live broadcast room feature information contained in the follow-up content, the live broadcast room feature information including the live broadcast room name, anchor user name and anchor user avatar; outputting and displaying the live broadcast room feature information in a graphical user interface, and in response to the live broadcast room entry instruction executed by the live broadcast room feature information, loading the live broadcast room page corresponding to the live broadcast room feature information.

2. The method according to claim 1, characterized in that After the step of acquiring the corresponding advertisement type tag from the content database and pushing the corresponding content with the advertisement type tag to the target user terminal according to the advertisement type tag of the advertisement material, the following post-processing steps are performed by the server: Monitoring user behavior data of the target user terminal in the live broadcast platform, and updating comprehensive return ratio information of the advertisement type corresponding to the advertisement type tag of the advertisement material contained in the advertisement message according to the user behavior data; According to the latest updated comprehensive return ratio information for the advertisement type, the ranking position of the advertisement type in the advertisement type ranking list is updated. The advertisement type ranking list contains advertisement types that are sorted according to their comprehensive return ratio information.

3. The method according to claim 1, characterized in that The step of inputting a plurality of advertising materials into an advertisement classification model that has been trained to convergence and obtaining advertisement type labels output by the advertisement classification model to which the respective advertising materials belong includes the following steps performed by the advertisement classification model: Pre-formatting the advertising material; Calling a feature extraction model that has been trained to convergence to extract a feature vector of the advertising creative; A label classification model that has been trained to convergence is called to calculate the probability that the feature vector corresponds to each advertisement type label, and the advertisement type label with the highest probability is used as the advertisement type label to which the advertisement material belongs.

4. The method according to claim 1, wherein The step of obtaining user characteristic information corresponding to the target user terminal, obtaining advertising materials with advertising type tags matching the user characteristic information from the advertising database, and generating an advertising message containing the advertising materials and pushing it to the target user terminal includes the following steps performed by the server: Acquire user characteristic information corresponding to the target user terminal from a user database, and determine the user type corresponding to the target user terminal based on the user characteristic information; querying the advertisement database for one or more advertisement type tags corresponding to the user type, and obtaining advertisement materials corresponding to each of the advertisement type tags and stored in the advertisement database; An advertisement message including each of the advertisement materials is generated and pushed to the target user terminal.

5. The method according to claim 1, wherein The step of responding to the platform entry event acting on the target user terminal, determining the advertising material in response to the target user terminal entering the live broadcast platform, and acquiring, based on the advertising type tag corresponding to the advertising material, the accompanying content with the advertising type tag from the content database and pushing it to the target user terminal includes the following steps executed by the server: In response to a platform entry event triggered by the target user terminal through any advertising material in the advertising message, obtaining an advertising material identifier corresponding to the platform entry event; Determine the advertising material pointed to by the advertising material identifier, and query the advertising type label corresponding to the advertising material stored in the advertising database; Acquire one or more follow-up contents with the advertisement type tag from the content database, and push the follow-up contents to the target user terminal.

6. A live broadcast platform information push device, characterized in that: It includes: An advertising material classification module is configured to input a plurality of advertising materials into an advertising classification model that has been trained to convergence, obtain advertising type labels output by the advertising classification model to which the advertising materials belong, and store the advertising materials and their advertising type labels in an advertising database; The acceptance content classification module is used to input multiple acceptance contents existing in the live broadcast platform into the advertisement classification model, obtain the advertisement type labels of the acceptance contents output by the advertisement classification model, and store the acceptance contents and their advertisement type labels in the content database in correspondence; An advertising message push module is configured to respond to an advertising push request from a target user terminal, obtain user characteristic information corresponding to the target user terminal, obtain advertising materials with advertising type tags matching the user characteristic information from the advertising database, and generate an advertising message containing the advertising materials and push it to the target user terminal; The takeover content push module is used to respond to the platform entry event acting on the target user terminal, determine the advertising material in response to the target user terminal entering the live broadcast platform, and according to the advertising type label corresponding to the advertising material, obtain the takeover content with the advertising type label from the content database and push it to the target user terminal, so that the target user terminal executes: receiving the takeover content pushed by the server, parsing the live broadcast room feature information contained in the takeover content, the live broadcast room feature information including the live broadcast room name, anchor user name and anchor user avatar; outputting and displaying the live broadcast room feature information in a graphical user interface, and loading the live broadcast room page corresponding to the live broadcast room feature information in response to the live broadcast room entry instruction executed by the live broadcast room feature information.

7. The device according to claim 6, characterized in that The content push module includes: an advertisement identification determination submodule, configured to respond to a platform entry event triggered by the target user terminal through any advertisement material in the advertisement message, and obtain an advertisement material identification corresponding to the platform entry event; An advertisement label determination submodule, configured to determine the advertisement material pointed to by the advertisement material identifier, and query the advertisement type label corresponding to the advertisement material stored in the advertisement database; The follow-up content acquisition submodule is used to acquire one or more follow-up contents with the advertisement type tag from the content database, and push the follow-up contents to the target user terminal.

8. An electronic device comprising a central processing unit and a memory, characterized in that: The central processing unit is configured to call and run a computer program stored in the memory to execute the steps of the method according to any one of claims 1 to 5.

9. A non-volatile storage medium, characterized in that: It stores a computer program implemented according to the method described in any one of claims 1 to 5 in the form of computer-readable instructions, and when the computer program is called and executed by a computer, the steps included in the method are executed.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images