Article recommendation method, device and equipment based on live broadcast, medium and program product
By identifying target events in the live broadcast data and pushing recommended items, the problem of cumbersome purchase paths in the live broadcast platform is solved, the purchase efficiency and targetedness are improved, and the interactive experience in the live broadcast room is improved.
Patent Information
- Application Number
- CN202310664571.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2023-06-06
- Publication Date
- 2024-12-06
AI Technical Summary
In the live broadcast platform, the process of viewers needing to manually log in, search and purchase items is cumbersome, resulting in inefficient purchases.
By obtaining live broadcast data, identifying target events and pushing corresponding recommended items to live broadcast accounts, simplifying the purchase process and realizing item recommendations and pushing.
It improves the efficiency and pertinence of item purchases, reduces the operation steps of the audience, and enhances the interactive experience in the live broadcast room.
Smart Images

Figure CN119090572A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to computer technology, and in particular to a method, device, equipment, medium and program product for recommending items based on live broadcast. Background Art
[0002] With the development of computer technology and live broadcast technology, more and more live broadcast platforms can broadcast various scenes (such as games, football matches, etc.). During the live broadcast process, in order to increase the audience's enthusiasm for interaction in the live broadcast room, relevant technologies have embedded in the live broadcast platform an item purchase page for purchasing items (such as game props or jerseys worn by players, etc.) involved in the live broadcast scene (such as games or football matches, etc.).
[0003] When purchasing items, viewers need to log in to the item purchase page first, then search for items that can be purchased on the item purchase page, and finally pay for the items that can be purchased. It can be seen that this purchase method has a long operation path and low purchase efficiency. Summary of the invention
[0004] The embodiments of the present application provide a method, device, electronic device, computer-readable storage medium and computer program product for recommending items based on live broadcast, which can improve the pertinence of item recommendations and the efficiency of item purchases.
[0005] The technical solution of the embodiment of the present application is implemented as follows:
[0006] The present application embodiment provides a method for recommending items based on live broadcast, including:
[0007] Acquire first live broadcast data of a first scene broadcast live on a live broadcast platform;
[0008] When the first live broadcast data includes data corresponding to a target event, determining a recommended item corresponding to the target event based on the data corresponding to the target event;
[0009] The recommended items are sent to the live broadcast platform, so that the live broadcast platform pushes the recommended items to a live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0010] The present application embodiment provides a method for recommending items based on live broadcast, including:
[0011] Displaying a live broadcast interface for live broadcasting of the first scene on a live broadcast platform, and displaying live broadcast content of the live broadcast in the live broadcast interface;
[0012] When the live broadcast content includes content corresponding to the target event, the recommended items corresponding to the target event are obtained from the live broadcast center, and the recommended items are pushed to the live broadcast account compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0013] The present application embodiment provides an item recommendation device based on live broadcast, including:
[0014] An acquisition module, used to acquire first live broadcast data of a first scene broadcast live on a live broadcast platform;
[0015] a determination module, configured to determine, when the first live broadcast data includes data corresponding to a target event, a recommended item corresponding to the target event based on the data corresponding to the target event;
[0016] The sending module is used to send the recommended items to the live broadcast platform, so that the live broadcast platform pushes the recommended items to the live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0017] In the above scheme, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the device also includes: a first binding module, used to receive a first binding request sent by the live broadcast platform, the first binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the scene account in the first scene; in response to the first binding request, establish a first binding relationship between the live broadcast account and the scene account; return the established first binding relationship to the live broadcast platform, so that when the live broadcast platform determines that the live broadcast account is compatible with the recommended item based on the first binding relationship, the recommended item is pushed to the live broadcast account.
[0018] In the above scheme, the first binding request carries the binding data corresponding to the live broadcast account, and the binding data is used to determine whether the live broadcast account is compatible with the recommended item; wherein, the binding data includes at least one of the following: the account data of the live broadcast account, and the note data of the live broadcast account for at least one of the concerned characters or concerned items in the first scene.
[0019] In the above scheme, the device also includes: other scene recommendation module, which is used to obtain second live broadcast data for live broadcasting of a second scene in the live broadcast platform, and the second scene is different from the first scene; when the second live broadcast data contains data corresponding to other events, based on the data corresponding to other events, other recommended items corresponding to the other events are determined, and when it is determined through the first binding relationship that the other recommended items are compatible with the live broadcast account, the other recommended items are pushed to the live broadcast account.
[0020] In the above scheme, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the device also includes: a second binding module, used to receive a second binding request sent by the live broadcast platform, the second binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the specific role in the first scene; in response to the second binding request, establish a second binding relationship between the live broadcast account and the specific role; return the established second binding relationship to the live broadcast platform, so that the live broadcast platform determines that the recommended item is compatible with the live broadcast account based on the second binding relationship, and pushes the recommended item to the live broadcast account when there is data of the target event corresponding to the specific role in the first live broadcast data.
[0021] In the above scheme, the device also includes: a binding switching module, used to receive a role switching request sent by the live broadcast platform, the role switching request is used to request that the specific role bound to the live broadcast account be switched to other roles; in response to the role switching request, the second binding relationship is released, and a third binding relationship is established between the live broadcast account and the other role; the established third binding relationship is returned to the live broadcast platform, so that the live broadcast platform, based on the third binding relationship, pushes recommended items corresponding to the target event to the live broadcast account when there is data of the target event corresponding to the other role in the first live broadcast data.
[0022] In the above scheme, before determining the recommended items corresponding to the target event based on the data of the corresponding target event, the device also includes: a first event determination module, used to obtain the component data corresponding to the target event, and match the first live broadcast data with the component data to obtain a matching result; when the matching result indicates that the matching degree between the first live broadcast data and the component data exceeds a matching degree threshold, it is determined that the first live broadcast data contains data corresponding to the target event.
[0023] In the above scheme, before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the device also includes: a second event determination module, which is used to determine that the first live broadcast data contains data corresponding to the target event when the first live broadcast data contains a target sentence for describing the target event and the number of occurrences of the target sentence reaches a number threshold.
[0024] In the above scheme, before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the device also includes: a third event determination module, which is used to perform event recognition on the interactive data when the first live broadcast data includes interactive data for the first scene, and obtain a corresponding recognition event; compare the recognition event with the target event, and when the recognition event is successfully compared with the target event, determine that the first live broadcast data contains data corresponding to the target event.
[0025] In the above scheme, the determination module is also used to obtain the live broadcast screen corresponding to the data of the corresponding target event; perform object recognition on the live broadcast screen to obtain at least one object contained in the live broadcast screen; and determine the recommended item corresponding to the target event based on the at least one object contained in the live broadcast screen.
[0026] In the above scheme, the determination module is also used to, when the number of items contained in the live broadcast screen is at least two, determine the at least two items as recommended items corresponding to the target event; or, determine the recommendation priority of each of the items, and determine the item among the at least two items whose recommendation priority is higher than the target priority as the recommended item corresponding to the target event.
[0027] In the above scheme, the device also includes: a priority determination module, which is used to determine the recommendation priority of the item through the following processing: obtaining the influencing parameters used to influence the recommendation priority of the item, wherein the influencing parameters include at least one of the following: the contribution of the item to the target event, the historical interaction data corresponding to the item; based on the influencing parameters, calling the neural network model to perform recommendation priority prediction processing to obtain the recommendation priority of the item; wherein the neural network model is trained through item samples, the influencing parameters corresponding to the item samples, and the recommendation priority annotated by the item samples.
[0028] In the above scheme, the determination module is also used to determine the associated items corresponding to the item based on at least one item contained in the live broadcast screen, and use the associated items as recommended items corresponding to the target event; wherein the similarity between the associated items and the items contained in the live broadcast screen exceeds a similarity threshold, or the matching degree between the associated items and the items contained in the live broadcast screen exceeds a matching threshold.
[0029] The embodiment of the present application also provides an item recommendation device based on live broadcast, including:
[0030] A display module, used to display a live broadcast interface for live broadcasting of the first scene on the live broadcast platform, and to display the live broadcast content of the live broadcast in the live broadcast interface;
[0031] The push module is used to obtain recommended items corresponding to the target event from the live broadcast center when the live broadcast content contains content corresponding to the target event, and push the recommended items to the live broadcast account compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0032] In the above scheme, after the recommended item is pushed to the live broadcast account that is compatible with the recommended item, the device also includes: a payment module, which is used to call the item payment protocol in the live broadcast platform when receiving a purchase operation for the recommended item from the live broadcast account, and deduct the payment resources of the live broadcast account for the recommended item according to the item payment protocol.
[0033] An embodiment of the present application provides an electronic device, including:
[0034] A memory for storing computer executable instructions or computer programs;
[0035] The processor is used to implement the live broadcast-based item recommendation method provided in the embodiment of the present application when executing the computer executable instructions or computer programs stored in the memory.
[0036] An embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions or a computer program for implementing the live broadcast-based item recommendation method provided in the embodiment of the present application when executed by a processor.
[0037] An embodiment of the present application provides a computer program product, including a computer program or computer executable instructions. When the computer program or computer executable instructions are executed by a processor, the live broadcast-based item recommendation method provided in the embodiment of the present application is implemented.
[0038] The embodiments of the present application have the following beneficial effects:
[0039] By applying the embodiment of the present application, during the live broadcast of the first scene on the live broadcast platform, when there is relevant content corresponding to the target event in the live broadcast content, the recommended items corresponding to the target event are immediately sent to the live broadcast platform, and the recommended items are pushed to the live broadcast account in the live broadcast platform through the live broadcast platform. That is, the live broadcast account does not need to manually select the recommended items, and the recommended items can be received for purchase by the live broadcast account, which shortens the purchase path of the items and can improve the purchase efficiency of the items in the live broadcast room; in addition, the live broadcast platform pushes the recommended items to the live broadcast account that is compatible with the recommended items, avoiding the disturbance of the live broadcast account by irrelevant recommended items, improving the targetedness of the item push, and is conducive to improving the conversion rate of the items. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 1 is a schematic diagram of the architecture of a live broadcast-based item recommendation system 100 provided in an embodiment of the present application;
[0041] Figure 2 is a schematic diagram of the structure of an electronic device 500 provided in an embodiment of the present application;
[0042] Figure 3 It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0043] Figure 4A It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0044] Figure 4B It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0045] Figure 4C It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0046] Figure 5 It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0047] Figure 6 It is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application;
[0048] Figure 7A-7B is a schematic diagram of displaying recommended items provided in an embodiment of the present application;
[0049] Figures 8A-8B This is a schematic diagram showing the display of purchase reminder information provided in an embodiment of the present application;
[0050] Fig. 9 This is a schematic diagram of account binding provided in an embodiment of the present application;
[0051] Fig.10 This is a schematic diagram of the account binding process provided by an embodiment of the present application;
[0052] Fig.11 This is a schematic diagram of account binding provided in an embodiment of the present application;
[0053] Fig.12 This is a schematic diagram of the item recommendation method based on live broadcast proposed in an embodiment of the present application. DETAILED DESCRIPTION
[0054] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.
[0055] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.
[0056] In the following description, the terms "first\second..." are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second..." can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.
[0057] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.
[0058] The embodiments of the present application provide a method, device, electronic device, computer-readable storage medium, and computer program product for recommending items based on live broadcast, which can meet the diverse information acquisition needs of users. The exemplary application of the electronic device provided by the embodiments of the present application is described below. The electronic device provided by the embodiments of the present application can be implemented as various types of user terminals such as laptop computers, tablet computers, desktop computers, set-top boxes, mobile devices (e.g., mobile phones, portable music players, personal digital assistants, dedicated messaging devices, portable gaming devices), smart phones, smart speakers, smart watches, smart TVs, and vehicle-mounted terminals, and can also be implemented as servers. Below, exemplary applications when the electronic device is implemented as a terminal will be described.
[0059] See also Figure 1 , Figure 1 It is an architectural diagram of a live broadcast-based item recommendation system 100 provided in an embodiment of the present application. To support an exemplary application, a terminal (terminal 400-1 and terminal 400-2 are shown as examples) is connected to a server 200 via a network 300. The network 300 may be a wide area network or a local area network, or a combination of the two.
[0060] In some embodiments, the terminal may be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, a car terminal, etc., but is not limited thereto. The terminal is provided with a client with a live broadcast function (which may be referred to as a live broadcast platform), such as an education client, an online conference client, an instant messaging client, a live broadcast client, etc. The server 200 is a background server corresponding to the client, such as a live broadcast middle station for controlling and managing the live broadcast platform. It may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The terminal and the server may be directly or indirectly connected via wired or wireless communication, which is not limited in the embodiments of the present application.
[0061] In actual applications, when a user participates in a live broadcast of a first scene (such as a game or a ball game, etc.) through a client (or a live broadcast platform) in a terminal, the terminal displays a live broadcast interface for the first scene in the live broadcast platform, displays the live broadcast content in the live broadcast interface, and sends the first live broadcast data corresponding to the live broadcast content to the server 200 (or the live broadcast center) in real time; the server 200 detects the first live broadcast data, and when the first live broadcast data contains data corresponding to a target event, based on the data corresponding to the target event, determines a recommended item corresponding to the target event, and sends the recommended item to the live broadcast platform, and pushes the recommended item to the terminal corresponding to the live broadcast account that matches the recommended item through the live broadcast platform for display, so that the live broadcast account can purchase the recommended item.
[0062] See also Figure 2 , Figure 2 is a schematic diagram of the structure of an electronic device 500 provided in an embodiment of the present application, taking the electronic device 500 as Figure 1 For example, the terminal in Figure 2 The electronic device 500 shown includes: at least one processor 510, a memory 550, at least one network interface 520 and a user interface 530. The various components in the electronic device 500 are coupled together via a bus system 540. It is understood that the bus system 540 is used to achieve connection and communication between these components. In addition to the data bus, the bus system 540 also includes a power bus, a control bus and a status signal bus. However, for the sake of clarity, the bus system 540 is not described in detail. Figure 2 Various buses are labeled as bus system 540 .
[0063] The processor 510 can be an integrated circuit chip with signal processing capabilities, such as a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc., where the general-purpose processor can be a microprocessor or any conventional processor, etc.
[0064] The memory 550 includes a volatile memory or a non-volatile memory, and may also include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), and the volatile memory may be a random access memory (RAM). The memory 550 described in the embodiment of the present application is intended to include any suitable type of memory. The memory 550 optionally includes one or more storage devices that are physically far away from the processor 510.
[0065] In some embodiments, the memory 550 can store data to support various operations, examples of which include programs, modules, and data structures, or a subset or superset thereof, as exemplarily described below.
[0066] The operating system 551 includes system programs for processing various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic businesses and processing hardware-based tasks; a network communication module 552 is used to reach other electronic devices via one or more (wired or wireless) network interfaces 520. Exemplary network interfaces 520 include: Bluetooth, Wireless Compatibility Certification (WiFi), and Universal Serial Bus (USB), etc.
[0067] In some embodiments, the live broadcast-based item recommendation device provided in the embodiments of the present application can be implemented in software. The live broadcast-based item recommendation device provided in the embodiments of the present application can be provided as various software embodiments, including various forms including applications, software, software modules, scripts or codes. Figure 2 A live broadcast-based item recommendation device 555 stored in a memory 550 is shown, which may be software in the form of a program or plug-in, and includes a series of modules, including an acquisition module 5551, a determination module 5552, and a sending module 5553. These modules are logical, and therefore may be arbitrarily combined or further split according to the functions implemented. The functions of each module will be described below.
[0068] In other embodiments, the device provided in the embodiments of the present application can be implemented in hardware. As an example, the device provided in the embodiments of the present application can be a processor in the form of a hardware decoding processor, which is programmed to execute the live broadcast-based item recommendation method provided in the embodiments of the present application. For example, the processor in the form of a hardware decoding processor can adopt one or more application-specific integrated circuits (Application Specific Integrated Circuit, ASIC), digital signal processors (Digital Signal Processor, DSP), programmable logic devices (Programmable Logic Device, PLD), complex programmable logic devices (Complex Programmable Logic Device, CPLD), field programmable gate arrays (Field-Programmable Gate Array, FPGA) or other electronic components.
[0069] In some embodiments, the terminal or server can implement the live broadcast-based item recommendation method provided in the embodiment of the present application by running various computer executable instructions or computer programs. For example, computer executable instructions can be microprogram-level commands, machine instructions or software instructions. The computer program can be a native program or software module in the operating system; it can be a native application (APPlication, APP), that is, a program that needs to be installed in the operating system to run, such as a live broadcast APP, an instant messaging APP; it can also be a small program, that is, a program that only needs to be downloaded to a browser environment to run; it can also be a small program that can be embedded in any APP. In short, the above-mentioned computer executable instructions can be instructions in any form, and the above-mentioned computer program can be an application, module or plug-in in any form.
[0070] As mentioned above, the method for recommending items based on live broadcast provided in the embodiment of the present application can be implemented by various types of electronic devices, for example, Figure 1 The terminal in the command prompt can also be executed separately by Figure 1 The terminal and the server 200 (i.e. the live broadcasting station) in the system are executed in coordination. Figure 1 The server (i.e., the live broadcasting middle station) in the embodiment of the present application independently executes the method for recommending items based on live broadcasting as an example. Figure 3 , Figure 3 is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application, which will be combined with Figure 3 The steps shown are explained.
[0071] Step 101: The live broadcasting station obtains first live broadcasting data of a first scene on a live broadcasting platform.
[0072] Here, in actual application, when a user participates in the live broadcast of the first scene (such as a game or a ball game, etc.) through the client (or live broadcast platform) in the terminal, the terminal displays the live broadcast interface of the first scene in the live broadcast platform, displays the live broadcast content of the live broadcast in the live broadcast interface, and sends the first live broadcast data corresponding to the live broadcast content to the live broadcast middle station in real time, wherein the first live broadcast data can reflect the real-time live broadcast content of the first scene, such as when the first scene is a game, the first live broadcast data can be the equipment data of the virtual props or virtual skills equipped by the virtual objects in the game in the game, or the attack data of the equipped virtual props or virtual skills attacking the enemy, or the special effect data of the attack special effects displayed by the enemy, etc.; for example, when the first scene is a ball game, the first live broadcast data can be the sports data of the players participating in the ball game kicking and passing on the court, or the equipment data of the jerseys or shoes worn by the players, etc. When the live broadcast middle station receives the first live broadcast data, it can detect the first live broadcast data in real time and perform subsequent operations based on the detection results.
[0073] Step 102: When the first live broadcast data includes data corresponding to the target event, based on the data corresponding to the target event, determine a recommended item corresponding to the target event.
[0074] Among them, the target event refers to the wonderful event (which can be considered as a general term for all wonderful events) corresponding to the highlight moment (also known as the peak moment or wonderful moment) in the first scene. For example, when the first scene is a game, the target event can be an event in which the player uses high-end operations (such as a killing streak operation) or kills the enemy. For example, when the first scene is a football match, the target event can be an event in which the player performs very well in the game, such as a goal event, winning the game, breaking a personal record or breaking a world record.
[0075] In some embodiments, see Figure 4A , Figure 4A is a flow chart of a method for recommending items based on live broadcast provided in an embodiment of the present application. Figure 3 Before step 102 shown, the live broadcasting station can also determine that the first live broadcasting data contains data corresponding to the target event through steps 104A to 105A:
[0076] In step 104A, component data corresponding to the target event is obtained, and the first live broadcast data is matched with the component data to obtain a matching result.
[0077] In actual applications, the component data corresponding to the target events in different scenes are different, and the component data corresponding to different target events in the same scene are also different. The component data corresponding to the target events in the scenes can be preset. For example, when the first scene is a game and the target event is a triple kill operation, the component data of the target event is the relevant data of the triple kill operation; for another example, when the first scene is a ball game and the target event is a goal event, the component data of the target event is the relevant data of the player kicking the ball and the kicked ball moving on the court until it enters the goal.
[0078] When determining whether the target event data exists in the first live broadcast data, the first live broadcast data may be matched with the component data corresponding to the target event (such as calculating the similarity value between the two, such as the cosine similarity value) to obtain a matching result, so as to determine whether the target event data exists in the first live broadcast data based on the matching result.
[0079] In step 105A, when the matching result indicates that the matching degree between the first live broadcast data and the component data exceeds a matching degree threshold, it is determined that the first live broadcast data contains data corresponding to the target event.
[0080] Here, when the matching result indicates that the degree of match between the first live broadcast data and the component data corresponding to the target event exceeds the matching threshold (such as the similarity value or the cosine similarity value exceeds the similarity threshold), it is determined that the first live broadcast data contains data corresponding to the target event; otherwise, when the matching result indicates that the degree of match between the first live broadcast data and the component data does not exceed the matching threshold (such as the similarity value or the cosine similarity value does not exceed the similarity threshold), it is determined that the first live broadcast data does not contain data corresponding to the target event.
[0081] In some embodiments, see Figure 4B , Figure 4B is a flow chart of a method for recommending items based on live broadcast provided in an embodiment of the present application. Figure 3 Before step 102 shown, the live broadcasting station may also determine in step 104B that the first live broadcasting data contains data corresponding to the target event:
[0082] In step 104B, when the first live broadcast data includes a target sentence for describing a target event and the number of occurrences of the target sentence reaches a number threshold, it is determined that the first live broadcast data includes data corresponding to the target event.
[0083] Here, in actual applications, the target event may have a corresponding description statement (i.e., target statement). For example, when the target event is a triple kill operation in a game, the description statement corresponding to the target event may be "xx triple kill"; when the target event is a goal event in a football game, the description statement corresponding to the target event may be "a goal is scored", and so on.
[0084] When the first live broadcast data includes the host's explanation data for the first scene, and there is a target sentence describing the target event in the explanation data, and the number of occurrences of the target sentence exceeds the number threshold (which can be set according to the actual application, such as 3), it can be determined that the first live broadcast data contains data corresponding to the target event. For example, when the first scene is a game, the target event is a triple kill operation, and the corresponding target sentence is "triple kill", if the first live broadcast data contains "triple kill" or "successful triple kill", and the number of occurrences reaches 3 times, it is determined that the first live broadcast data contains data corresponding to the target event.
[0085] In some embodiments, see Figure 4C , Figure 4C is a flow chart of a method for recommending items based on live broadcast provided in an embodiment of the present application. Figure 3 Before step 102 shown, the live broadcasting station can also determine that the first live broadcasting data contains data corresponding to the target event through steps 104C to 105C:
[0086] In step 104C, when the first live broadcast data includes interactive data for the first scene, event recognition is performed on the interactive data to obtain a corresponding recognition event.
[0087] Among them, the interactive data refers to the interactive data of the host or audience in the live broadcast room for the first scene in the live broadcast, such as the number of likes, comment data (number of comments and comment content), virtual gift data, etc. The interactive data can reflect the progress or actual situation of the first scene in the live broadcast to a certain extent. For example, when there is a wonderful moment in the first scene, the number of likes at the wonderful moment is much greater than the number of likes at non-wonderful moments.
[0088] In practical applications, usually, the interactive data corresponding to target events in different scenes may be different, and the interactive data (such as comment content) corresponding to different target events in the same scene may also be different. For example, when the first scene is a game, if the target event is a triple kill, the interactive data is the comment content about the triple kill; if the target event is the release of a flying skill, the interactive data is the comment content about the release of a flying skill; for another example, when the first scene is a football game, if the target event is a high-quality goal, the interactive data is the comment content about the goal; if the target event is a high-quality block, the interactive data is the comment content about the block.
[0089] When the first live broadcast data includes interactive data for the first scene, event recognition can be performed on the interactive data. When there are multiple types of interactive data, event recognition can be performed in combination with the multiple interactive data. For example, if the number of likes corresponding to the first live broadcast data broadcast at the current moment is significantly higher than the number of likes at other moments, it can be determined that the first live broadcast data broadcast at the current moment may contain data corresponding to exciting events. Then, keyword extraction is further performed on the comment content corresponding to the first live broadcast data, and the corresponding recognition event is determined based on the extracted keywords. For example, when the first scene is a game, the comment content is mostly related to keywords such as "continuous kill" and "triple kill", then the recognition event can be determined to be a continuous kill event; for example, when the first scene is a ball game, the comment content is mostly related to keywords such as "three-point shot" and "goal", then the recognition event can be determined to be a goal event.
[0090] In step 105C, the identified event is compared with the target event, and when the identified event is successfully compared with the target event, it is determined that the first live broadcast data contains data corresponding to the target event.
[0091] Here, the target event can be regarded as a wonderful event pre-set in the first scene. When the recognized event is successfully matched with the target event pre-set in the first scene (for example, if the similarity between the recognized event and the target event reaches more than 95%, it is considered that the two are successfully matched), it is determined that the first live broadcast data contains data corresponding to the target event.
[0092] In some embodiments, see Figure 5 , Figure 5 is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application. Figure 3 In step 102 shown in FIG. 1 , based on the data corresponding to the target event, the recommended items corresponding to the target event are determined by Figure 5 Steps 1021-1022 in the above are used to implement:
[0093] In step 1021, a live broadcast screen corresponding to the data corresponding to the target event is obtained, and object recognition is performed on the live broadcast screen to obtain at least one object contained in the live broadcast screen.
[0094] Here, the live broadcast center obtains the live broadcast screen corresponding to the data of the corresponding target event from the live broadcast platform (that is, the video frame image corresponding to the current live broadcast video content), and performs object recognition on the live broadcast screen, such as inputting the live broadcast screen into the trained object recognition model to obtain at least one object contained in the live broadcast screen. For example, when the first scene is a game, the at least one object contained in the live broadcast screen may be a virtual prop, virtual skill, etc. used by the player to trigger the target event (such as a triple kill); for another example, when the first scene is a football game, the at least one object contained in the live broadcast screen may be a jersey, sneakers, etc. worn by the player who triggered the target event (such as a goal).
[0095] In step 1022, based on at least one item included in the live broadcast screen, a recommended item corresponding to the target event is determined.
[0096] In some embodiments, Figure 5 Step 1022 shown can be implemented in the following manner: when the number of items contained in the live broadcast screen is at least two, at least two items are determined as recommended items corresponding to the target event; or, the recommendation priority of each item is determined, and the item with a recommendation priority higher than the target priority among the at least two items is determined as the recommended item corresponding to the target event.
[0097] Here, when the live broadcast screen contains only one item, the identified item can be directly used as the recommended item corresponding to the target event; when the live broadcast screen contains multiple items (two or more), the identified multiple items can be used as recommended items corresponding to the target event (i.e., multiple recommended items), and the live broadcast center sends all the recommended items to the live broadcast platform, and the live broadcast platform pushes each recommended item to the live broadcast account that matches the recommended item, such as pushing recommended item 1 to the corresponding live broadcast account 1, and pushing recommended item 2 to the corresponding live broadcast account 2; in this way, when the first live broadcast data contains data of the same target event, the recommended items corresponding to the target event received by the terminals on different live broadcast account sides may also be different, thereby improving the targeted push of recommended items. In addition, while the live broadcast platform pushes each recommended item to the live broadcast account that is compatible with the recommended item, it can also push other recommended items to the corresponding live broadcast account. For example, if the compatibility between recommended item 1 and live broadcast account 1 is higher than that between recommended item 2 and live broadcast account 1, the live broadcast platform will push recommended item 1 to live broadcast account 1 and also push recommended item 2 to live broadcast account 1. Live broadcast account 1 can choose the recommended item to purchase from them, thereby increasing the diversity of choices.
[0098] In some embodiments, when the live broadcast screen contains multiple items (two or more), the live broadcast center can also filter out specific items from the multiple items as recommended items corresponding to the target event, such as filtering out items with the highest recommendation priority as recommended items corresponding to the target event. For example, when the live broadcast screen contains items 1, 2, and 3, if the recommendation priorities of these three items decrease in sequence, item 1 will be used as the recommended item corresponding to the target event. The live broadcast center will send the recommended item (i.e., item 1) to the live broadcast platform, and push the recommended item (i.e., item 1) to the live broadcast account that is compatible with the recommended item through the live broadcast platform. This will improve the targeted push of recommended items while also helping to improve the conversion rate of recommended items.
[0099] In some embodiments, the live broadcast station can determine the recommendation priority of an item by the following processing: obtaining influencing parameters for influencing the recommendation priority of an item, wherein the influencing parameters include at least one of the following: the contribution of the item to the target event, the historical interaction data corresponding to the item; based on the influencing parameters, calling a neural network model to perform recommendation priority prediction processing to obtain the recommendation priority of the item; wherein the neural network model is trained through item samples, the influencing parameters corresponding to the item samples, and the recommendation priority annotated by the item samples.
[0100] Among them, the contribution of an item to a target event refers to the size of the power contributed by the item to triggering the target event. For example, when the first scene is a game and the target event is a killing streak event, virtual item 1 (i.e. item 1) provides 80% of the power to the killing streak event, and virtual item 2 (i.e. item 2) provides 10% of the power to the killing streak event. It can be considered that the contribution of virtual item 1 (i.e. item 1) to the killing streak event is greater than the contribution of virtual item 2 (i.e. item 1) to the killing streak event. Generally, the greater the contribution of an item to a target event, the greater its influence on the recommendation priority. When the influence of other influencing parameters on the recommendation priority is fixed, the greater the contribution of an item to the target event, the greater the recommendation priority corresponding to the item.
[0101] The historical interactive data corresponding to an item refers to the interactive data such as the visits and purchase volume of the item by the live broadcast account after the item is pushed to the live broadcast platform in the historical live broadcast of the first scene. Generally, the greater the historical visits and / or purchase volume corresponding to the item, the greater its influence on the recommendation priority. When the influence of other influencing parameters on the recommendation priority is fixed, the greater the historical visits and / or purchase volume corresponding to the item, the greater the recommendation priority corresponding to the item. After obtaining the influencing parameters used to influence the recommendation priority of the item, the live broadcast center performs prediction processing in the neural network model trained with the influencing parameters to obtain the recommendation priority of the item, which can make the prediction more accurate.
[0102] Before applying the neural network model, it is necessary to train the initial neural network model, and then put the trained neural network model into use to combine the influencing parameters of the items and predict the recommendation priority of the items through artificial intelligence technology. Among them, the neural network model is trained by the influencing parameters corresponding to the item samples and the recommendation priority annotated by the item samples. For example, based on the item samples, the contribution of the item samples to the target event and the recommendation priority annotated by the item samples, the initial neural network model is called for prediction processing to obtain the predicted recommendation priority. After determining the value of the loss function of the neural network model through the predicted recommendation priority and the annotated recommendation priority, it can be determined whether the value of the loss function exceeds the preset threshold. When the value of the loss function exceeds the preset threshold, the error signal of the neural network model is determined based on the loss function, the error information is back-propagated in the neural network model, and the model parameters of each layer are updated during the propagation process.
[0103] Among them, the embodiments of the present application do not limit the model structure of the neural network model. For example, the neural network model can be a convolutional neural network, a deep neural network, etc.; it is not limited to the form of the loss function, for example, it can be a cross entropy loss function, an L2 loss function, etc.
[0104] Here, back propagation is explained. The training sample data is input into the input layer of the neural network model, passes through the hidden layer, and finally reaches the output layer and outputs the result. This is the forward propagation process of the neural network model. Since there is an error between the output result of the neural network model and the actual result, the error between the output result and the actual value is calculated, and the error is backpropagated from the output layer to the hidden layer until it propagates to the input layer. In the process of back propagation, the value of the model parameter is adjusted according to the error, that is, the loss function is constructed according to the error between the output result and the actual value, and the partial derivatives of the loss function with respect to the model parameters are calculated layer by layer to generate the gradient of the loss function with respect to the model parameters of each layer. Since the direction of the gradient indicates the direction of error expansion, the gradient of the model parameters is inverted and summed with the original parameters of each layer model. The summed result is used as the updated model parameters of each layer, thereby reducing the error caused by the model parameters, and the above process is continuously iterated until convergence.
[0105] In some embodiments, Figure 5 Step 1022 shown may be implemented in the following manner: based on at least one object included in the live broadcast screen, determining an associated item corresponding to the item, and using the associated item as a recommended item corresponding to the target event.
[0106] Here, when the live broadcast screen contains at least one item, associated items related to the item can be recommended as recommended items corresponding to the target event, wherein the similarity between the associated items and the items contained in the live broadcast screen exceeds a similarity threshold, or the matching degree between the associated items and the items contained in the live broadcast screen exceeds a matching threshold.
[0107] For example, when the first scene is a game, assuming that the item included in the live broadcast screen is virtual prop 1, virtual prop 2 whose similarity with virtual prop 1 exceeds the similarity threshold (that is, virtual prop 2 is an associated item) can be used as a recommended item corresponding to the target event; or, when virtual prop 1 is a virtual firearm, virtual prop 3 whose matching degree with the virtual prop exceeds the matching threshold (such as a magazine or bullet that matches virtual prop 1) can be used as a recommended item corresponding to the target event.
[0108] In actual applications, the items contained in the live broadcast screen and the related items corresponding to the items can also be used as recommended items corresponding to the target event, and the items contained in the live broadcast screen and the related items corresponding to the items can be sent to the live broadcast platform together, and the items contained in the live broadcast screen and the related items corresponding to the items can be pushed to the corresponding live broadcast account through the live broadcast platform; in this way, the items pushed to the live broadcast account are not only the items that trigger the target event, but also other items related to the items that trigger the target event, so that the live broadcast account has more choices when purchasing items, which is conducive to improving the conversion rate of recommended items.
[0109] Step 103: Send the recommended items to the live broadcast platform, so that the live broadcast platform pushes the recommended items to the live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0110] In some embodiments, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the live broadcast center may receive a first binding request sent by the live broadcast platform, the first binding request being used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the scene account in the first scene; in response to the first binding request, a first binding relationship between the live broadcast account and the scene account is established; and the established first binding relationship is returned to the live broadcast platform, so that when the live broadcast platform determines that the live broadcast account is compatible with the recommended item based on the first binding relationship, the recommended item is pushed to the live broadcast account.
[0111] Here, all live broadcast accounts participating in the live broadcast of the first scene on the live broadcast platform can send a first binding request to the live broadcast center through the live broadcast platform before or during the live broadcast of the first scene, wherein the first binding request carries the account identifier of the live broadcast account to be bound and the scene identifier of the first scene. After receiving the first binding request, the live broadcast center establishes a first binding relationship between the live broadcast account and the scene account in response to the first binding request, and returns the established first binding relationship to the live broadcast platform.
[0112] During the live broadcast of the first scene on the live broadcast platform, if the first live broadcast data contains data corresponding to the target event, the live broadcast center obtains the recommended items corresponding to the target event, and sends the recommended items to the live broadcast platform. After the live broadcast platform receives the recommended items corresponding to the target event sent by the live broadcast center, it determines the live broadcast account that is compatible with the recommended items based on the first binding relationship between all live broadcast accounts and the first scene, and pushes the pushed items to the live broadcast account that is compatible with the recommended items.
[0113] In some embodiments, the first binding request carries binding data corresponding to the live broadcast account, wherein the binding data is used to determine whether the live broadcast account is compatible with the recommended item; wherein the binding data includes at least one of the following: account data of the live broadcast account, and note data of the live broadcast account for at least one of the concerned characters or concerned items in the first scene.
[0114] Here, the account data of the live broadcast account includes account level, account preference, etc., which can affect the item push rate of the live broadcast platform for the live broadcast account. In actual applications, recommended items can have a certain recommendation level. When sending the recommended items to the live broadcast platform, the live broadcast middle station also sends the recommendation level of the recommended items to the live broadcast platform. The live broadcast platform can push the recommended items to the corresponding live broadcast account according to the recommendation level of the recommended items, such as pushing the recommended items to the live broadcast account corresponding to the account level that matches the recommendation level (if the recommendation level of the recommended item is not lower than the account level of the live broadcast account, it is considered that the recommendation level is compatible with the account level, and the recommended item is considered to be compatible with the live broadcast account), or push the recommended items to the live broadcast account corresponding to the account preference that matches the recommendation level.
[0115] The remark information of the live broadcast account includes at least one of the concerned characters or concerned items for the first scene in the remark of the live broadcast account. For example, when the first scene is a game, the remark information of the live broadcast account may be "Favorite virtual character A" or "Expect to get virtual prop B". Then, when the target event is a wonderful event corresponding to virtual character A, it can be considered that the recommended item corresponding to the target event is compatible with the live broadcast account; or, when the recommended item corresponding to the target event is virtual prop B, it can be considered that the recommended item corresponding to the target event is compatible with the live broadcast account. For another example, when the first scene is a ball game, the remark information of the live broadcast account may be "I want the same pair of sneakers as my favorite star C". Then, when the target event is a wonderful event corresponding to star C and the recommended item is the sneakers D that star C is wearing, it can be considered that the recommended item corresponding to the target event (i.e., sneakers D) is compatible with the live broadcast account.
[0116] Through the above method, when establishing the first binding relationship between the live broadcast account in the live broadcast platform and the scene account of the first scene, the binding data corresponding to the live broadcast account carried by the first binding request is recorded to quickly determine whether the live broadcast account and the recommended item are compatible based on the binding data.
[0117] In some embodiments, after establishing the first binding relationship, the live broadcast center can also obtain second live broadcast data for a second scene on the live broadcast platform, where the second scene is different from the first scene; when the second live broadcast data contains data corresponding to other events, other recommended items corresponding to other events are determined based on the data corresponding to other events, and when it is determined through the first binding relationship that other recommended items are compatible with the live broadcast account, the other recommended items are pushed to the live broadcast account.
[0118] Here, after the live broadcast platform establishes the first binding relationship between the live broadcast account in the live broadcast platform and the scene account of the first scene, it can be compatible with the interface protocols of different scenes, that is, the live broadcast account and the first scene are bound once in the live broadcast platform, and other scenes (such as a second scene different from the first scene) are subsequently broadcast on the live broadcast platform without the need to bind to the live broadcast account for the second time. The live broadcast platform can directly push other recommended items corresponding to the second scene to the live broadcast account for purchase by the live broadcast account based on the first binding relationship between the live broadcast account and the first scene, thereby reducing the tediousness of item purchase and improving the user's purchasing experience.
[0119] In some embodiments, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the live broadcast center may also receive a second binding request sent by the live broadcast platform, where the second binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the specific character in the first scene; in response to the second binding request, a second binding relationship between the live broadcast account and the specific character is established; and the established second binding relationship is returned to the live broadcast platform, so that the live broadcast platform determines that the recommended item is compatible with the live broadcast account based on the second binding relationship, and pushes the recommended item to the live broadcast account when there is data of the target event corresponding to the specific character in the first live broadcast data.
[0120] Here, when the second binding request is used to request the establishment of a binding relationship between a live broadcast account in the live broadcast platform and a specific character in the first scene, such as requesting the establishment of a binding relationship between a live broadcast account A and a virtual character B (i.e., a specific character) in the first scene, the live broadcast center responds to the second binding request to establish a second binding relationship between the live broadcast account A and the virtual character B (i.e., the specific character), wherein the second binding relationship includes the established binding relationship between the live broadcast account A and the scene account of the first scene and the binding relationship between the live broadcast account A and the virtual character B in the first scene, and the second binding relationship is sent to the live broadcast platform; in this way, the first scene is broadcast live on the live broadcast platform. In the process, if the first live broadcast data contains data corresponding to the target event, and the target event is an event related to the virtual character B in the first scene, after the live broadcast center obtains the recommended item corresponding to the target event, it sends the information that the target event is an event related to the virtual character B in the first scene together with the recommended item to the live broadcast platform. After the live broadcast platform receives the information sent by the live broadcast center, based on the information that the target event is an event related to the virtual character B in the first scene and the second binding relationship, it can be determined that the recommended item corresponding to the target event is compatible with the live broadcast account A, so the recommended item can be pushed to the live broadcast account A to improve the targeted nature of the item push.
[0121] In some embodiments, the live broadcast center may also receive a role switching request sent by the live broadcast platform, where the role switching request is used to request that a specific role bound to the live broadcast account be switched to another role; in response to the role switching request, the second binding relationship is released, and a third binding relationship is established between the live broadcast account and the other role; the established third binding relationship is returned to the live broadcast platform, so that the live broadcast platform, based on the third binding relationship, pushes recommended items corresponding to the target event to the live broadcast account when there is data of the target event corresponding to other characters in the first live broadcast data.
[0122] Here, the role switching request is used to request to switch a specific role bound to the live broadcast account to another role. For example, the second binding relationship is bound to the relationship between the live broadcast account A and the virtual character B (i.e., the specific role) in the first scene. The role switching request is used to request to switch the virtual character B (i.e., the specific role) bound to the live broadcast account A to the virtual character C (i.e., the other role). In response to the role switching request, the live broadcast center releases the second binding relationship between the live broadcast account A and the virtual character B (i.e., the specific role) in the first scene, and establishes a third binding relationship between the live broadcast account A and the virtual character C (i.e., the other role), and sends the established third binding relationship to the live broadcast platform. In this way, in the live broadcast platform During the live broadcast of the first scene, if the first live broadcast data contains data corresponding to other events and the target event is an event related to the virtual character C in the first scene, after the live broadcast center obtains the recommended item corresponding to the target event, it sends the information that the target event is an event related to the virtual character C in the first scene together with the recommended item to the live broadcast platform. After the live broadcast platform receives the information sent by the live broadcast center, based on the information that the target event is an event related to the virtual character C in the first scene and the third binding relationship, it can be determined that the recommended item corresponding to the target event is compatible with the live broadcast account A. Therefore, the recommended item can be pushed to the live broadcast account A to improve the targeted nature of the item push.
[0123] Through the above method, after the live broadcast center determines the recommended items corresponding to the target event, it can send the determined recommended items to the live broadcast platform, and push the recommended items to the terminal corresponding to the live broadcast account compatible with the recommended items through the live broadcast platform, that is, the push page for the recommended items is displayed in the terminal corresponding to the live broadcast account compatible with the recommended items. The push page can display the detailed information, price and purchase entrance of the recommended items, and the live broadcast account can purchase the recommended items through the purchase entrance in the push page. It should be noted that after the live broadcast account purchases the recommended items, it can use the purchased recommended items by itself, or give or sell the purchased recommended items to other live broadcast accounts.
[0124] It should be noted that when a user enters the live broadcast platform through an instant messaging client, the live broadcast account provided in the embodiment of the present application can be the user's login account in the instant messaging client.
[0125] In some embodiments, Figure 1 The terminal (i.e., the live broadcast platform) in the embodiment of the present application independently executes the method for recommending items based on live broadcast provided in the embodiment of the present application as an example, see Figure 6 , Figure 6 is a flowchart of a method for recommending items based on live broadcast provided in an embodiment of the present application, which will be combined with Figure 6 The steps shown are explained.
[0126] Step 201: Display a live broadcast interface for broadcasting a first scene on a live broadcast platform, and display the live broadcast content on the live broadcast interface.
[0127] Here, in actual application, when the user participates in the live broadcast of the first scene (such as a game or a ball game, etc.) through the client (or live broadcast platform) in the terminal, the terminal displays the live broadcast interface for the first scene in the live broadcast platform, displays the live broadcast content in the live broadcast interface, and sends the live broadcast content or the first live broadcast data corresponding to the live broadcast content to the live broadcast center in real time. After the live broadcast center receives the live broadcast content or the first live broadcast data corresponding to the live broadcast content, it determines in real time whether the live broadcast content contains the content of the corresponding target event, or determines whether the first live broadcast data contains the data of the corresponding target event.
[0128] Step 202: When the live broadcast content includes content corresponding to the target event, obtain the recommended items corresponding to the target event from the live broadcast center, and push the recommended items to the live broadcast account compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0129] Here, when the live broadcast content contains content corresponding to the target event, the live broadcast platform pulls the recommended items corresponding to the target event from the live broadcast platform, and pushes the recommended items to the live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0130] It can be understood that the method for judging whether the live broadcast content contains content corresponding to the target event is consistent with the above-mentioned method for judging whether the first live broadcast data contains data corresponding to the target event. The method for determining recommended items and the method for judging whether the recommended items are compatible with the live broadcast account are the same as the above-mentioned methods and will not be repeated here.
[0131] In some embodiments, after the terminal pushes the recommended item to the live broadcast account that is compatible with the recommended item, when receiving a purchase operation for the recommended item from the live broadcast account, the terminal calls the item payment protocol in the live broadcast platform and deducts the payment resources of the live broadcast account for the recommended item according to the item payment protocol.
[0132] See also Figure 7A-7B , Figure 7A-7B : is a display schematic diagram of recommended items provided in an embodiment of the present application. When the live broadcast platform pushes the recommended items to the terminal corresponding to the live broadcast account compatible with the recommended items, a push page for the recommended items can be displayed in the terminal corresponding to the live broadcast account compatible with the recommended items. The push page can display the detailed information, payment method, price and purchase entrance (i.e., immediate purchase control) of the recommended items, among which the payment methods include point payment and cash payment, such as Fig. 7A The payment method is point payment. Figure 7BThe payment method is cash.
[0133] After the live broadcast account selects a payment method on the push page, the live broadcast platform responds to the payment operation triggered by the purchase entry, calls the item payment protocol in the live broadcast platform, and determines according to the item payment protocol that the current live broadcast account has the purchase authority (such as the remaining resources in the live broadcast account are greater than or equal to the resources consumed by purchasing the recommended items, and the recommended items have not reached the maximum purchase limit), then uses the selected payment method to pay, deducts the payment resources of the live broadcast account for the recommended items, and displays a prompt message of successful purchase, such as Fig. 8A If the current live broadcast account does not have the purchase authority (for example, the remaining resources in the live broadcast account are less than the resources consumed by purchasing the recommended items, or the recommended items have reached the maximum purchase limit), a purchase failure prompt message will be displayed, such as Figure 8B As shown, Figures 8A-8B This is a schematic diagram of displaying purchase reminder information provided in an embodiment of the present application.
[0134] It should be noted that after a live broadcast account purchases a recommended item, the purchased recommended item can be viewed in the scene account corresponding to the live broadcast account in the first scene. The purchased recommended item can be used by the live broadcast account himself, or given or sold to other live broadcast accounts.
[0135] It is understandable that in the embodiments of the present application, when user information, account data, note data and other related data are involved, when the embodiments of the present application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.
[0136] Below, an exemplary application of the embodiment of the present application in an actual application scenario will be described. Take the live broadcast scene as a game (that is, the first scene is a game), the target event as a highlight moment event, and the recommended item as a virtual prop as an example. In order to ensure one-time access to the live broadcast platform and realize the push function (or sales function) of the highlight moment virtual props of multiple games, the live broadcast middle station provides a standardized access service. The live broadcast middle station is responsible for compatibility with the props sales interface protocols of different games and the game highlight moment data. First, the live broadcast middle station aligns the props payment agreement and the props issuance agreement with the game application (that is, the game client), and then aligns the game highlight moment push agreement. Finally, the live broadcast middle station will encapsulate the props payment agreement (including game token payment and cash payment), and integrate the encapsulated props payment agreement in the live broadcast middle station prop shop to facilitate the live broadcast platform to call. At the same time, the live broadcast center will synchronize the highlight moment data to the live broadcast platform. At the highlight moment of the event, the live broadcast platform will open the prop shop of the live broadcast center and display the prop push page corresponding to the highlight moment event on the live broadcast platform (that is, the push page of the recommended items mentioned above). The virtual props corresponding to the highlight moment event will be displayed on the prop push page for purchase by the live broadcast accounts participating in the live broadcast on the live broadcast platform.
[0137] See also Fig. 9 , Fig. 9 This is a schematic diagram of account binding provided in an embodiment of the present application. In order to enhance the user's purchasing experience and reduce the complexity of purchasing, the live broadcast platform provides a game account binding service, that is, to bind the user's live broadcast account in the live broadcast platform with the game account in the game. After the user is bound, when there is a highlight moment event in the game live broadcast, the live broadcast platform can be used to pull up the props purchase. For subsequent props purchase activities, the same live broadcast account does not need to log in twice.
[0138] See also Fig.10 , Fig.10 This is a schematic diagram of the account binding process provided by an embodiment of the present application. The method includes:
[0139] Step 301: The live broadcast platform sends a first binding request to the live broadcast platform, wherein the first binding request is used to request to establish a binding relationship between the first live broadcast account and the game account corresponding to the target game.
[0140] In actual applications, the anchor can bind the live broadcast account with the game account based on the account binding page of the live broadcast platform, such as displaying at least one game that has not been bound and the binding account control corresponding to the game on the account binding page. When the anchor wants to bind with the game account of the target game, the binding account control corresponding to the target game can be triggered. The live broadcast platform responds to the triggering operation of the binding account control corresponding to the target game and displays at least one live broadcast account for selection, such as the first live broadcast account and the second live broadcast account. It should be noted that the at least one live broadcast account for selection displayed is an account that the anchor has logged in to the live broadcast platform in the past. When the anchor enters the live broadcast platform through the instant messaging client, the live broadcast account can be the login account of the anchor in the instant messaging client, such as the first live broadcast account is the login account of the anchor in the instant messaging client 1, and the second live broadcast account is the login account of the anchor in the instant messaging client 2. When there is no live broadcast account for selection (that is, the anchor has not logged in to the account in the live broadcast platform in the past), it is necessary to add a new live broadcast account with the authorization of the anchor for account binding.
[0141] When the anchor selects the first live broadcast account, the live broadcast platform can generate and send a first binding request for requesting to establish a binding relationship between the first live broadcast account and the game account corresponding to the target game to the live broadcast center in response to the selection operation.
[0142] Step 302: The live broadcast platform responds to the first binding request by establishing a first binding relationship between the first live broadcast account and the game account corresponding to the target game, and returns the established first binding relationship to the live broadcast platform.
[0143] The live broadcasting platform has developed a game account binding intermediate page, which is invoked by the anchor in the live broadcasting platform. For example, the live broadcasting platform sends the first binding request triggered by the anchor to the live broadcasting platform. Since the first binding request carries the first live broadcast account of the anchor to be bound and the game account of the target game, the live broadcasting platform transmits the anchor's live broadcast account on the live broadcasting platform side to the account binding intermediate page of the live broadcasting platform. At this point, the account binding intermediate page of the live broadcasting platform has relevant information of the live broadcasting platform. When the live broadcasting platform receives the first binding request sent by the live broadcasting platform, it responds to the first binding request and establishes the first binding relationship between the first live broadcast account and the game account corresponding to the target game, that is, the mapping relationship between the anchor's live broadcast account on the live broadcasting platform and the game account in the target game is implemented in the live broadcasting platform.
[0144] After the live broadcast center binds the anchor and the target game (i.e., establishes a first binding relationship between the first live broadcast account and the game account corresponding to the target game), the binding information (i.e., the first binding relationship) is returned to the live broadcast platform. After the live broadcast platform receives the binding information (i.e., the first binding relationship), based on the first binding relationship, a switching account control is displayed at the associated position of the target game in the account binding page, wherein the switching account control is used to switch the first live broadcast account bound to the target game to another live broadcast account.
[0145] Step 303: The live broadcast platform sends an account switching request to the live broadcast platform, wherein the account switching request is used to request to switch the live broadcast account bound to the game account corresponding to the target game.
[0146] When the host wants to switch the live broadcast account bound to the game account of the target game, the account switching control corresponding to the target game can be triggered. The live broadcast platform responds to the triggering operation of the account switching control and displays at least one live broadcast account available for selection, such as the second live broadcast account, the third live broadcast account, and the fourth live broadcast account. When the host selects the target live broadcast account (such as the third live broadcast account), the live broadcast platform responds to the selection operation and generates and sends an account switching request to the live broadcast center to request switching the first live broadcast account bound to the game account corresponding to the target game to the target live broadcast account.
[0147] Step 304: In response to the account switching request, the live broadcast platform switches the first live broadcast account bound to the game account corresponding to the target game in the first binding relationship to the target live broadcast account, and returns the switching result to the live broadcast platform.
[0148] Through the above method, the first binding relationship is switched from the relationship between the first live broadcast account and the game account corresponding to the target game to the relationship between the target live broadcast account and the game account corresponding to the target game.
[0149] Step 305: The live broadcast platform sends a second binding request to the live broadcast platform, wherein the second binding request is used to request to establish a binding relationship between the target live broadcast account and a specific character in the target game.
[0150] In actual applications, after binding the host's live broadcast account with the game account of the target game, if the host has not yet bound the virtual character in the target game, the role binding control can be displayed at the associated position of the target game in the account binding page. The role binding control is used to establish a binding relationship between the target host account and the game character in the target game; when the host wants to bind with the game character in the target game, the role binding control corresponding to the target game can be triggered, and the live broadcast platform responds to the triggering operation of the role binding control corresponding to the target game, and displays at least one game character available for selection in the target game, such as game character A and game character B. It can be understood that the at least one game character available for selection can be a character that the host has controlled and played in the target game, or it can be any game character involved in the target game.
[0151] When the anchor chooses to bind a specific character (such as game character A), the live broadcast platform can generate and send a second binding request to the live broadcast platform in response to the selection operation, requesting to establish a binding relationship between the target live broadcast account and the specific character in the target game.
[0152] Step 306: The live broadcast platform responds to the second binding request, establishes a second binding relationship between the target live broadcast account and the specific character in the target game, and returns the established second binding relationship to the live broadcast platform.
[0153] After the live broadcast platform receives the second binding relationship, based on the second binding relationship, a character switching control is displayed at the associated position of the target game in the account binding page, wherein the character switching control is used to switch a specific character in the target game bound to the target live broadcast account.
[0154] Step 307: The live broadcast platform sends a role switching request to the live broadcast platform, wherein the role switching request is used to request to switch a specific role in the target game bound to the target live broadcast account.
[0155] When the anchor wants to switch to a specific character in the target game bound to the target live broadcast account, the corresponding character switching control of the target game can be triggered. The live broadcast platform responds to the trigger operation of the character switching control and displays at least one game character for selection, such as game character B and game character C. When the anchor selects other characters to be switched (such as game character C), the live broadcast platform responds to the selection operation and generates and sends a character switching request to the live broadcast center for requesting to switch the specific character in the target game bound to the target live broadcast account to other characters.
[0156] Step 308: In response to the role switching request, the live broadcast platform switches the specific role in the target game bound to the target live broadcast account in the second binding relationship to another role, and returns the switching result to the live broadcast platform.
[0157] Here, the second binding relationship is switched from the relationship between the target live broadcast account and the game character A in the target game to the relationship between the target live broadcast account and the game character C in the target game.
[0158] The product performance in the account binding page involved in the above steps 301 to 308 can be found in Fig.11 , Fig.11 This is an account binding schematic diagram provided in an embodiment of the present application. A target game that has not yet been bound is displayed in the account binding page in the live broadcast platform. Since the target game has not yet been bound to a live broadcast account or a virtual character, a binding account control and a binding role control are displayed next to the target game. After a first binding relationship between the target game and the first live broadcast account is established through the binding account control, the binding account control displayed next to the target game is transformed into an account switching control, and the first live broadcast account bound to the target game can be switched to other live broadcast accounts through the account switching control; after a second binding relationship between a specific character in the target game and the first live broadcast account is established through the binding role control, the binding role control displayed next to the target game is transformed into a role switching control, and the specific character bound to the first live broadcast account can be switched to other characters through the role switching control.
[0159] See also Fig.12 , Fig.12 : is a schematic diagram of a method for recommending items based on live broadcasting according to an embodiment of the present application, the method comprising:
[0160] In step 401, the target game pushes the highlight moment data to the live broadcast platform, that is, during the live broadcast of the target game on the live broadcast platform, the target game pushes the highlight moment data to the live broadcast platform when the highlight moment data occurs according to the actual situation of the live broadcast; in step 402, the live broadcast platform dynamically generates a push page for virtual props according to the highlight moment data, that is, the live broadcast platform renders the prop shop according to the highlight moment data, and dynamically generates a push page for virtual props (also called a prop sales page); in step 403, the highlight moment event is notified to the live broadcast platform during the live broadcast; in step 404, the live broadcast platform pulls the push page for virtual props from the live broadcast platform.
[0161] Taking the establishment of the second binding relationship between the target live broadcast account and the specific character in the target game as an example, if the live broadcast data contains data corresponding to the highlight moment event, and the highlight moment event is an event related to the specific character in the target game, the live broadcast center obtains the virtual props corresponding to the highlight moment event, and pulls up the protocol through the prop push page to send the information that the highlight moment event is a highlight moment event related to the specific character in the target game together with the virtual props to the live broadcast platform. After receiving the information sent by the live broadcast center, the live broadcast platform determines that the virtual props corresponding to the highlight moment event are compatible with the target live broadcast account, so it can push the virtual props to the terminal corresponding to the target live broadcast account for the target live broadcast account to purchase, so as to improve the targeted push of virtual props.
[0162] When the live broadcast platform pushes the virtual props to the terminal corresponding to the target live broadcast account, the push page of the virtual props in the live broadcast platform can be automatically pulled in the terminal corresponding to the target live broadcast account, and the push page for the virtual props can be displayed. The push page can display the details of the virtual props, payment methods, prices and purchase entrances (i.e., immediate purchase controls), etc. Among them, the payment methods include point payment and cash payment, such as Fig. 7A The payment method is point payment. Figure 7B The payment method is cash.
[0163] When the target live broadcast account selects a payment method on the push page, the live broadcast platform responds to the payment operation triggered by the purchase entry, calls the props payment protocol in the live broadcast platform, and when it is determined that the target live broadcast account has the purchase authority according to the props payment protocol, the payment is made using the selected payment method, the payment resources of the target live broadcast account for the recommended items are deducted, and a prompt message indicating a successful purchase is displayed, such as Fig. 8A If the target live broadcast account does not have the purchase authority, a purchase failure prompt message will be displayed, such as Figure 8B shown.
[0164] In actual applications, when the payment method is cash payment, the live broadcast center can determine whether the target live broadcast account has the purchase authority by querying the remaining balance in the target live broadcast account. For example, when the remaining balance in the target live broadcast account is greater than or equal to the funds consumed to purchase virtual props, and the virtual props have not yet reached the maximum purchase limit, the target live broadcast account is considered to have the purchase authority. In this case, the amount consumed to purchase virtual props can be directly deducted from the remaining balance in the target live broadcast account to purchase the virtual props; if the remaining balance in the target live broadcast account is less than the funds consumed to purchase virtual props, or the virtual props have reached the maximum purchase limit, it is considered that the target live broadcast account does not have the purchase authority, and the virtual props cannot be successfully purchased at this time.
[0165] When the payment method is point payment (i.e., game token payment), the live broadcast center calls the game token query interface to query the game tokens in the game account bound to the target live broadcast account to determine whether the target live broadcast account has the purchase authority. For example, when the queried game tokens are greater than or equal to the game tokens consumed to purchase virtual props, and the virtual props have not yet reached the maximum purchase limit, the target live broadcast account is considered to have the purchase authority. In this case, the game token deduction interface can be called to deduct the game tokens consumed to purchase virtual props from the game tokens in the game account corresponding to the target live broadcast account to purchase the virtual props. The purchased virtual props can be viewed in the game account corresponding to the target live broadcast account in the target game, and the target game can return the deduction status and balance to the live broadcast center; when the queried game tokens are less than the game tokens consumed to purchase virtual props, or the virtual props have reached the maximum purchase limit, it is considered that the target live broadcast account does not have the purchase authority, and the virtual props cannot be successfully purchased at this time.
[0166] Through the above method, the needs of the audience of the live broadcast platform to quickly purchase game props in the live broadcast room can be realized, the linkage between the live broadcast platform and the game can be effectively solved, more active users can be provided for the game, the operational capabilities of the live broadcast platform can be increased, and high-quality props can be recommended to players in combination with the real-time live broadcast scene. At the same time, a general component with single access and multiple uses can be provided for the live broadcast platform, which reduces development costs and improves the overall experience of the audience of the live broadcast platform.
[0167] So far, the live broadcast-based item recommendation method provided by the embodiment of the present application has been described in combination with the exemplary application and implementation of the electronic device provided by the embodiment of the present application. The following will continue to describe the cooperation of various modules in the live broadcast-based item recommendation device 555 provided by the embodiment of the present application to implement the live broadcast-based item recommendation scheme.
[0168] The acquisition module 5551 is used to acquire first live broadcast data for broadcasting a first scene on a live broadcast platform; the determination module 5552 is used to determine the recommended items corresponding to the target event based on the data of the corresponding target event when the first live broadcast data contains data of the corresponding target event; the sending module 5553 is used to send the recommended items to the live broadcast platform, so that the live broadcast platform pushes the recommended items to a live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0169] In some embodiments, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the device also includes: a first binding module, used to receive a first binding request sent by the live broadcast platform, the first binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the scene account in the first scene; in response to the first binding request, establish a first binding relationship between the live broadcast account and the scene account; return the established first binding relationship to the live broadcast platform, so that when the live broadcast platform determines that the live broadcast account is compatible with the recommended item based on the first binding relationship, the recommended item is pushed to the live broadcast account.
[0170] In some embodiments, the first binding request carries binding data corresponding to the live broadcast account, and the binding data is used to determine whether the live broadcast account is compatible with the recommended item; wherein, the binding data includes at least one of the following: account data of the live broadcast account, and note data of the live broadcast account for at least one of the concerned characters or concerned items in the first scene.
[0171] In some embodiments, the device also includes: other scene recommendation modules, used to obtain second live broadcast data for live broadcasting a second scene in the live broadcast platform, and the second scene is different from the first scene; when the second live broadcast data contains data corresponding to other events, based on the data corresponding to other events, other recommended items corresponding to the other events are determined, and when it is determined through the first binding relationship that the other recommended items are compatible with the live broadcast account, the other recommended items are pushed to the live broadcast account.
[0172] In some embodiments, before obtaining the first live broadcast data for broadcasting the first scene on the live broadcast platform, the device also includes: a second binding module, used to receive a second binding request sent by the live broadcast platform, the second binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the specific role in the first scene; in response to the second binding request, establish a second binding relationship between the live broadcast account and the specific role; return the established second binding relationship to the live broadcast platform, so that the live broadcast platform determines that the recommended item is compatible with the live broadcast account based on the second binding relationship, and pushes the recommended item to the live broadcast account when there is data of the target event corresponding to the specific role in the first live broadcast data.
[0173] In some embodiments, the device also includes: a binding switching module, used to receive a role switching request sent by the live broadcast platform, the role switching request is used to request that the specific role bound to the live broadcast account be switched to other roles; in response to the role switching request, the second binding relationship is released, and a third binding relationship is established between the live broadcast account and the other role; the established third binding relationship is returned to the live broadcast platform, so that the live broadcast platform, based on the third binding relationship, pushes recommended items corresponding to the target event to the live broadcast account when there is data of the target event corresponding to the other character in the first live broadcast data.
[0174] In some embodiments, before determining the recommended items corresponding to the target event based on the data of the corresponding target event, the device also includes: a first event determination module, used to obtain component data corresponding to the target event, and match the first live broadcast data with the component data to obtain a matching result; when the matching result indicates that the matching degree between the first live broadcast data and the component data exceeds a matching degree threshold, it is determined that the first live broadcast data contains data corresponding to the target event.
[0175] In some embodiments, before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the device also includes: a second event determination module, which is used to determine whether the first live broadcast data contains data corresponding to the target event when the first live broadcast data contains a target sentence for describing the target event and the number of occurrences of the target sentence reaches a number threshold.
[0176] In some embodiments, before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the device also includes: a third event determination module, which is used to perform event recognition on the interactive data to obtain a corresponding recognition event when the first live broadcast data includes interactive data for the first scene; compare the recognition event with the target event, and when the recognition event is successfully compared with the target event, determine that the first live broadcast data contains data corresponding to the target event.
[0177] In some embodiments, the determination module is further used to obtain a live broadcast screen corresponding to the data of the corresponding target event; perform object recognition on the live broadcast screen to obtain at least one object contained in the live broadcast screen; and determine a recommended item corresponding to the target event based on the at least one object contained in the live broadcast screen.
[0178] In some embodiments, the determination module is also used to, when the number of items contained in the live broadcast screen is at least two, determine the at least two items as recommended items corresponding to the target event; or, determine the recommendation priority of each of the items, and determine the item among the at least two items whose recommendation priority is higher than the target priority as the recommended item corresponding to the target event.
[0179] In some embodiments, the device also includes: a priority determination module, used to determine the recommendation priority of the item through the following processing: obtaining influencing parameters used to influence the recommendation priority of the item, wherein the influencing parameters include at least one of the following: the contribution of the item to the target event, the historical interaction data corresponding to the item; based on the influencing parameters, calling a neural network model to perform recommendation priority prediction processing to obtain the recommendation priority of the item; wherein the neural network model is trained through item samples, the influencing parameters corresponding to the item samples, and the recommendation priority annotated by the item samples.
[0180] In some embodiments, the determination module is further used to determine an associated item corresponding to the item based on at least one item contained in the live broadcast screen, and use the associated item as a recommended item corresponding to the target event; wherein the similarity between the associated item and the item contained in the live broadcast screen exceeds a similarity threshold, or the matching degree between the associated item and the item contained in the live broadcast screen exceeds a matching threshold.
[0181] The embodiment of the present application also provides an item recommendation device based on live broadcast, including:
[0182] A display module, used to display a live broadcast interface for live broadcasting of the first scene on the live broadcast platform, and to display the live broadcast content of the live broadcast in the live broadcast interface;
[0183] The push module is used to obtain recommended items corresponding to the target event from the live broadcast center when the live broadcast content contains content corresponding to the target event, and push the recommended items to the live broadcast account compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
[0184] In some embodiments, after the recommended item is pushed to a live broadcast account that is compatible with the recommended item, the device also includes: a payment module, which is used to call the item payment protocol in the live broadcast platform when a purchase operation for the recommended item is received from the live broadcast account, and deduct the payment resources of the live broadcast account for the recommended item according to the item payment protocol.
[0185] The embodiment of the present application provides a computer program product, which includes a computer program or a computer executable instruction, and the computer program or the computer executable instruction is stored in a computer-readable storage medium. The processor of the electronic device reads the computer executable instruction from the computer-readable storage medium, and the processor executes the computer executable instruction, so that the electronic device executes the above-mentioned live broadcast-based item recommendation method of the embodiment of the present application.
[0186] The embodiment of the present application provides a computer-readable storage medium storing computer-executable instructions, wherein the computer-executable instructions or computer programs are stored. When the computer-executable instructions or computer programs are executed by a processor, the processor will be caused to execute the xx method provided in the embodiment of the present application, for example, Figure 3 The item recommendation method based on live broadcast is shown.
[0187] In some embodiments, the computer-readable storage medium may be a memory such as FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface storage, optical disk, or CD-ROM; or it may be various devices including one or any combination of the above memories.
[0188] In some embodiments, computer executable instructions may be in the form of a program, software, software module, script or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine or other unit suitable for use in a computing environment.
[0189] As an example, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file storing other programs or data, such as in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files storing one or more modules, subroutines, or code portions).
[0190] As an example, computer executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed at multiple sites and interconnected by a communication network.
[0191] The above is only an embodiment of the present application and is not intended to limit the protection scope of the present application. Any modifications, equivalent substitutions and improvements made within the spirit and scope of the present application are included in the protection scope of the present application.
Claims
1. A method for recommending items based on live broadcasting. It is characterized in that The method comprises: Acquire first live broadcast data of a first scene broadcast live on a live broadcast platform; When the first live broadcast data includes data corresponding to a target event, determining a recommended item corresponding to the target event based on the data corresponding to the target event; The recommended items are sent to the live broadcast platform, so that the live broadcast platform pushes the recommended items to a live broadcast account that is compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
2. The method according to claim 1, It is characterized in that Before obtaining first live broadcast data for live broadcasting the first scene on the live broadcast platform, the method further includes: receiving a first binding request sent by the live broadcast platform, where the first binding request is used to request to establish a binding relationship between a live broadcast account in the live broadcast platform and a scene account in the first scene; In response to the first binding request, establishing a first binding relationship between the live broadcast account and the scene account; The first established binding relationship is returned to the live broadcast platform, so that when the live broadcast platform determines that the live broadcast account is compatible with the recommended item based on the first binding relationship, the recommended item is pushed to the live broadcast account.
3. The method according to claim 2, It is characterized in that The first binding request carries binding data corresponding to the live broadcast account, and the binding data is used to determine whether the live broadcast account is compatible with the recommended item; The binding data includes at least one of the following: account data of the live broadcast account, and remark data of the live broadcast account for at least one of the concerned characters or concerned items in the first scene.
4. The method according to claim 2, It is characterized in that The method further comprises: Acquire second live broadcast data of a second scene broadcast live on the live broadcast platform, where the second scene is different from the first scene; When the second live broadcast data includes data corresponding to other events, other recommended items corresponding to the other events are determined based on the data corresponding to the other events, and when it is determined through the first binding relationship that the other recommended items are compatible with the live broadcast account, the other recommended items are pushed to the live broadcast account.
5. The method according to claim 1, It is characterized in that Before obtaining first live broadcast data for live broadcasting the first scene on the live broadcast platform, the method further includes: receiving a second binding request sent by the live broadcast platform, where the second binding request is used to request to establish a binding relationship between the live broadcast account in the live broadcast platform and the specific role in the first scene; In response to the second binding request, establishing a second binding relationship between the live broadcast account and the specific role; The second binding relationship established is returned to the live broadcast platform, so that the live broadcast platform, based on the second binding relationship, determines that the recommended item is compatible with the live broadcast account when there is data of the target event corresponding to the specific character in the first live broadcast data, and pushes the recommended item to the live broadcast account.
6. The method according to claim 5, It is characterized in that The method further comprises: Receiving a role switching request sent by the live broadcast platform, wherein the role switching request is used to request to switch the specific role bound to the live broadcast account to another role; In response to the role switching request, releasing the second binding relationship, and establishing a third binding relationship between the live broadcast account and the other role; The established third binding relationship is returned to the live broadcast platform, so that the live broadcast platform, based on the third binding relationship, pushes the recommended items corresponding to the target event to the live broadcast account when there is data of the target event corresponding to the other character in the first live broadcast data.
7. The method according to claim 1, It is characterized in that Before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the method further includes: Acquire component data corresponding to the target event, and match the first live broadcast data with the component data to obtain a matching result; When the matching result indicates that the matching degree between the first live broadcast data and the component data exceeds a matching degree threshold, it is determined that the first live broadcast data contains data corresponding to the target event.
8. The method according to claim 1, It is characterized in that Before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the method further includes: When the first live broadcast data includes a target sentence for describing a target event, and the number of occurrences of the target sentence reaches a number threshold, it is determined that the first live broadcast data includes data corresponding to the target event.
9. The method according to claim 1, It is characterized in that Before determining the recommended items corresponding to the target event based on the data corresponding to the target event, the method further includes: When the first live broadcast data includes interactive data for the first scene, performing event recognition on the interactive data to obtain a corresponding recognition event; The identified event is compared with the target event, and when the identified event is successfully compared with the target event, it is determined that the first live broadcast data contains data corresponding to the target event.
10. The method according to claim 1, It is characterized in that The determining, based on the data corresponding to the target event, a recommended item corresponding to the target event, comprises: Obtain a live broadcast screen corresponding to the data corresponding to the target event; Performing object recognition on the live broadcast screen to obtain at least one object contained in the live broadcast screen; Based on at least one item included in the live broadcast screen, a recommended item corresponding to the target event is determined.
11. The method according to claim 10, It is characterized in that The determining, based on at least one item included in the live broadcast screen, a recommended item corresponding to the target event includes: When the number of items included in the live broadcast screen is at least two, determining the at least two items as recommended items corresponding to the target event; or, The recommendation priority of each of the items is determined, and the item with a recommendation priority higher than the target priority among the at least two items is determined as a recommended item corresponding to the target event.
12. The method according to claim 11, It is characterized in that The method further comprises: The recommendation priority of the items is determined by the following process: Acquire an influencing parameter for influencing the recommendation priority of the item, wherein the influencing parameter includes at least one of the following: the contribution of the item to the target event, and the historical interaction data corresponding to the item; Based on the influencing parameters, a neural network model is called to perform recommendation priority prediction processing to obtain the recommendation priority of the item; The neural network model is obtained by training through item samples, influencing parameters corresponding to the item samples, and recommendation priorities annotated by the item samples.
13. The method according to claim 10, It is characterized in that The determining, based on at least one item included in the live broadcast screen, a recommended item corresponding to the target event includes: Based on at least one item included in the live broadcast screen, determining an associated item corresponding to the item, and using the associated item as a recommended item corresponding to the target event; The similarity between the associated item and the item included in the live broadcast screen exceeds a similarity threshold, or the matching degree between the associated item and the item included in the live broadcast screen exceeds a matching threshold.
14. A method for recommending items based on live broadcast. It is characterized in that The method comprises: Displaying a live broadcast interface for live broadcasting of the first scene on a live broadcast platform, and displaying live broadcast content of the live broadcast in the live broadcast interface; When the live broadcast content includes content corresponding to the target event, the recommended items corresponding to the target event are obtained from the live broadcast center, and the recommended items are pushed to the live broadcast account compatible with the recommended items, so that the live broadcast account can purchase the recommended items.
15. The method of claim 14, It is characterized in that After pushing the recommended item to a live broadcast account compatible with the recommended item, the method further includes: When a purchase operation for the recommended item from the live broadcast account is received, the item payment protocol in the live broadcast platform is called, and the payment resources of the live broadcast account for the recommended item are deducted according to the item payment protocol.
16. An item recommendation device based on live broadcast, It is characterized in that The device comprises: An acquisition module, used to acquire first live broadcast data of a first scene broadcast live on a live broadcast platform; a determination module, configured to determine, when the first live broadcast data includes data corresponding to a target event, a recommended item corresponding to the target event based on the data corresponding to the target event; The sending module is used to send the recommended items to the live broadcast platform, so that the live broadcast platform pushes the recommended items to the live broadcast account compatible with the recommended items.
17. An item recommendation device based on live broadcast, It is characterized in that The device comprises: A display module, used to display a live broadcast interface for live broadcasting of the first scene on the live broadcast platform, and to display the live broadcast content of the live broadcast in the live broadcast interface; The push module is used to obtain recommended items corresponding to the target event from the live broadcast center when the live broadcast content contains content corresponding to the target event, and push the recommended items to the live broadcast account compatible with the recommended items.
18. An electronic device, It is characterized in that include: A memory for storing computer executable instructions or computer programs; A processor, used to implement the live broadcast-based item recommendation method as described in any one of claims 1 to 15 when executing the computer executable instructions or computer programs stored in the memory.
19. A computer-readable storage medium, It is characterized in that Computer executable instructions or computer programs are stored, and when the computer executable instructions or computer programs are executed by a processor, the live broadcast-based item recommendation method described in any one of claims 1 to 15 is implemented.
20. A computer program product comprising a computer program or computer executable instructions, It is characterized in that When the computer program or computer executable instructions are executed by a processor, the live broadcast-based item recommendation method described in any one of claims 1 to 15 is implemented.
Citation Information
Cited By
Information display method and device, electronic equipment and storage medium
CN120091149A
