Resource processing method and device for live broadcast program

Abstract

The embodiment of the invention provides a resource processing method and device for a live broadcast program, and the method comprises the steps: receiving a resource request which is submitted by a user terminal and is used for the resource interaction of an access user in the live broadcast program in a resource processing process, determining a resource interaction type of the resource request according to the resource interaction record of the access user, reading a resource interaction numerical value obtained and stored by calling a large language model through an intelligent agent to perform resource interaction recommendation under the condition that the resource interaction type is determined to be a preset interaction type, and returning the resource interaction numerical value to the user terminal, and finally, performing resource interaction processing according to the target resource numerical value submitted by the user terminal, thereby performing high-quality resource processing through the cooperation of the intelligent agent and the large language model.

Description

Technical Field

[0001] This document relates to the field of network live broadcast technology, and in particular to a resource processing method and device for a live broadcast program. Background Art

[0002] With the continuous development and promotion of the Internet, the application scope of various online services provided by the Internet is becoming wider and wider. For example, in the field of live streaming, users can not only watch live content but also interact with resources during the process of accessing live streaming. In this case, a method of purchasing resources online during the process of accessing live streaming has emerged and has gradually been accepted by users. However, as more and more resource institutions access this method and the diversity of users' needs for resource processing increases, how to improve user satisfaction and increase user retention rate has become the focus of attention of all parties. Summary of the Invention

[0003] One or more embodiments of this specification provide a resource processing method for a live broadcast program, comprising: receiving a resource request submitted by a user terminal for a user accessing the live broadcast program to interact with resources. Determining the resource interaction type of the resource request based on the resource interaction record of the accessing user. If the resource interaction type is a preset interaction type, reading a resource interaction value obtained and stored by an intelligent agent calling a large language model for resource interaction recommendation. Returning the resource interaction value to the user terminal, and performing resource interaction processing based on the target resource value submitted by the user terminal.

[0004] One or more embodiments of this specification provide another resource processing method for a live broadcast program, comprising: generating a resource request for resource interaction based on a resource interaction operation submitted by a user accessing the live broadcast program and submitting the request to a server; receiving and displaying a resource interaction value for the resource interaction returned by the server; the resource interaction value being obtained by an intelligent agent using a large language model to perform resource interaction recommendation; obtaining a target resource value selected by the user from the resource interaction value and submitting the request to the server for performing resource interaction processing based on the target resource value.

[0005] One or more embodiments of the present specification provide a resource processing device for a live broadcast program, comprising: a request receiving module configured to receive a resource request submitted by a user terminal for a visiting user to interact with resources in the live broadcast program. An interaction type determination module configured to determine the resource interaction type of the resource request based on the resource interaction record of the visiting user. A value reading module configured to read the resource interaction value obtained and stored by the intelligent agent calling a large language model for resource interaction recommendation if the resource interaction type is a preset interaction type. A resource interaction module configured to return the resource interaction value to the user terminal and perform resource interaction processing based on the target resource value submitted by the user terminal.

[0006] One or more embodiments of the present specification provide another resource processing device for a live broadcast program, including: a request submission module, configured to generate a resource request for resource interaction based on the resource interaction operation submitted by the accessing user in the live broadcast program and submit it to the server. A value display module, configured to receive and display the resource interaction value for resource interaction returned by the server; the resource interaction value is obtained by the intelligent agent calling a large language model to perform resource interaction recommendation. A target value submission module, configured to obtain the target resource value selected by the accessing user from the resource interaction value and submit it to the server to perform resource interaction processing based on the target resource value.

[0007] One or more embodiments of the present specification provide a server, comprising: a processor; and a memory configured to store computer-executable instructions, wherein when the computer-executable instructions are executed, the processor: receives a resource request submitted by a user terminal for accessing a user to interact with resources in a live broadcast program. Determine the resource interaction type of the resource request based on the resource interaction record of the accessing user. If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the intelligent agent calling a large language model for resource interaction recommendation. Return the resource interaction value to the user terminal, and perform resource interaction processing based on the target resource value submitted by the user terminal.

[0008] One or more embodiments of this specification provide a user terminal, comprising: a processor; and a memory configured to store computer-executable instructions, wherein when executed, the computer-executable instructions cause the processor to: generate a resource request for resource interaction based on a resource interaction operation submitted by an accessing user in a live broadcast program and submit the request to a server. Receive and display a resource interaction value for resource interaction returned by the server; the resource interaction value is obtained by an intelligent agent calling a large language model to perform resource interaction recommendation. Obtain a target resource value selected by the accessing user from the resource interaction value and submit the request to the server to perform resource interaction processing based on the target resource value.

[0009] One or more embodiments of this specification provide a computer-readable storage medium for storing computer-executable instructions, which implement the following process when executed: receiving a resource request submitted by a user terminal for accessing a user to interact with resources in a live broadcast program. Determine the resource interaction type of the resource request based on the resource interaction record of the accessing user. If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the intelligent agent calling a large language model for resource interaction recommendation. Return the resource interaction value to the user terminal, and perform resource interaction processing based on the target resource value submitted by the user terminal.

[0010] One or more embodiments of this specification provide another computer-readable storage medium for storing computer-executable instructions, which implement the following process when executed: based on the resource interaction operation submitted by the accessing user in the live broadcast program, generate a resource request for resource interaction and submit it to the server. Receive and display the resource interaction value returned by the server for resource interaction; the resource interaction value is obtained by the intelligent agent calling a large language model to perform resource interaction recommendation. Obtain the target resource value selected by the accessing user in the resource interaction value and submit it to the server to perform resource interaction processing according to the target resource value. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate one or more embodiments of this specification or technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or descriptions of the prior art. Obviously, the drawings described below are only some embodiments described in this specification. Those skilled in the art can derive other drawings based on these drawings without inventive effort. Figure 1 A schematic diagram of an implementation environment for a resource processing method for a live broadcast program provided in one or more embodiments of this specification; Figure 2 A flowchart of a resource processing method for a live broadcast program provided in one or more embodiments of this specification; Figure 3 A flowchart of a resource processing method for a live broadcast program applied to a resource processing scenario provided by one or more embodiments of this specification; Figure 4 A flowchart of another method for processing resources of a live broadcast program provided in one or more embodiments of this specification; Figure 5 A schematic diagram of an embodiment of a resource processing device for a live broadcast program provided in one or more embodiments of this specification; Figure 6 A schematic diagram of another embodiment of a resource processing device for a live broadcast program provided in one or more embodiments of this specification; Figure 7 A schematic diagram of the structure of a server provided in one or more embodiments of this specification; Figure 8 A schematic diagram of the structure of a user terminal provided in one or more embodiments of this specification. DETAILED DESCRIPTION

[0012] In order to enable those skilled in the art to better understand the technical solutions in one or more embodiments of this specification, the technical solutions in one or more embodiments of this specification will be clearly and completely described below in conjunction with the drawings in one or more embodiments of this specification. Obviously, the described embodiments are only part of the embodiments of this specification, not all of the embodiments. Based on one or more embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of this document.

[0013] The resource processing method of the live broadcast program provided in one or more embodiments of this specification can be applied to the implementation environment of network live broadcast. Figure 1 , the implementation environment includes at least: Access user's user terminal 101 and server 102; The user terminal 101 may run a live broadcast program 101-1. The user terminal 101 may be used to access the live broadcast room and interact with the live broadcast, and cooperate with the server 102 to process resource interactions. The user terminal 102 may specifically be a mobile phone, a personal computer, a tablet computer, an e-book reader, a device for information interaction based on VR (Virtual Reality), an in-vehicle terminal, an IoT device, a wearable smart device, a laptop computer, a desktop computer, etc. The server 102 can run an intelligent agent 102-1 and a large language model 102-2. The server 102 is used to respond to resource requests for resource interaction submitted by the user terminal 101, and read the resource interaction value obtained by the intelligent agent 102-1 calling the large language model 102-2 for resource interaction recommendation, return the resource interaction value to the user terminal 101 and cooperate with the user terminal 101 to perform resource interaction processing; the server 102 can be a single server, or a server cluster composed of several servers, or one or more cloud servers in a cloud computing platform.

[0014] The implementation environment may also include the host's live broadcast terminal 103, which can be used to access the live broadcast room and interact with the live broadcast, and cooperate with the server 102 to perform corresponding processing during the live broadcast interaction; the live broadcast terminal 103 can specifically be a mobile phone, personal computer, tablet computer, e-book reader, equipment based on VR (Virtual Reality) for information interaction, vehicle-mounted terminal, IoT device, wearable smart device, laptop portable computer and desktop computer, etc.

[0015] In this implementation environment, when the user terminal 101 and the server 102 cooperate to perform resource interaction processing during the live broadcast, the user terminal 101 first generates a resource request for resource interaction based on the resource interaction operation submitted by the visiting user in the live broadcast program 101-1 and submits it to the server 102. The server 102 receives the resource request submitted by the user terminal 101 and determines the resource interaction type of the resource request based on the resource interaction record of the visiting user. When the resource interaction type is a preset interaction type, the server 102 reads the resource interaction value obtained and stored by the intelligent agent 102-1 calling the large language model 102-2 for resource interaction recommendation, and returns the resource interaction value to the user terminal 101. The user terminal 101 receives the resource interaction value and displays it, and submits it to the server 102 after obtaining the target resource value selected by the visiting user in the resource interaction value. Finally, the server 102 performs resource interaction processing based on the target resource value submitted by the user terminal 101, thereby realizing resource interaction processing based on the resource interaction value recommended to the user during the live broadcast.

[0016] One or more embodiments of a resource processing method for a live broadcast program provided in this specification are as follows: Reference Figure 2 The resource processing method of the live broadcast program provided in this embodiment specifically includes steps S202 to S208.

[0017] Step S202: receiving a resource request submitted by a user terminal for accessing resources for user interaction in a live broadcast program.

[0018] The live broadcast program described in this embodiment refers to a program that can be used for live broadcasting. The live broadcast program can specifically be a live broadcast subroutine, or can also be a service subroutine for live broadcasting in a resource service program. The live broadcast program can run in an application program for live broadcasting.

[0019] In this embodiment, the visiting user can interact with the host who is performing the live broadcast while accessing the live broadcast program. For example, the visiting user can donate resources to the host, or the visiting user can use resources to redeem virtual objects and donate the redeemed virtual objects to the host. Among them, in the process of the visiting user donating resources and / or redeeming resources to the host, in order to enable the visiting user to have sufficient resources for interaction, the visiting user can obtain resources by submitting resource interaction operations to perform resource interaction.

[0020] Among them, resource interaction refers to the interactive processing of resources by visiting users, specifically refers to the purchase or recharge of resources by visiting users in the live broadcast program; among them, resources refer to virtual resources, specifically refers to virtual resources that can be purchased, recharged, donated to the anchor and / or used in the live broadcast program.

[0021] During specific implementation, when an accessing user accesses online live broadcasting through a live broadcasting program, the accessing user may submit a resource interaction operation in the live broadcasting program through a user terminal. The user terminal generates and submits a resource request for resource interaction based on the resource interaction operation. Correspondingly, the resource request for resource interaction in the live broadcasting program submitted by the accessing user from the user terminal is received. Specifically, the resource interaction operation submitted by the accessing user may be a trigger operation submitted through the resource interaction interface of the live broadcasting program. In this case, the resource interaction operation includes a trigger operation for the resource interaction interface of the live broadcasting program.

[0022] Step S204: determining the resource interaction type of the resource request according to the resource interaction record of the accessing user.

[0023] In specific implementation, when receiving a resource request from a visiting user for resource interaction, in order to improve the satisfaction and retention rate of the visiting user in the subsequent resource processing process, and thereby promote the success rate of resource interaction, the resource interaction type of the resource request is determined based on the resource interaction record of the visiting user.

[0024] Among them, resource interaction records refer to records of resource interaction operations performed by accessing users in live broadcast programs. For example, resource interaction records may include the number of resource interactions and / or resource interaction time of accessing users. In addition, resource interaction records may also include other data related to resource interaction operations, such as resource interaction methods; the resource interaction type refers to the type of resource interaction operation performed by accessing users. For example, the resource interaction type may be the resource interaction type of the accessing user performing the resource interaction operation for the first time, or it may be the resource interaction type of the accessing user performing the resource interaction operation for a non-first time.

[0025] During the specific implementation process, in the process of determining the resource interaction type of the resource request based on the resource interaction record of the accessing user, whether the resource interaction type is the resource interaction type of the first resource interaction operation can be determined by detecting whether there is a resource interaction record of the accessing user, so that different resource values are returned to the accessing user in the subsequent resource processing process for different resource interaction types. In an optional implementation provided by this embodiment, determining the resource interaction type of the resource request based on the resource interaction record of the accessing user includes: Detect whether there is a resource interaction record of the live broadcast program in the resource flow record of the application program of the accessing user running the live broadcast program; If it exists, determine the resource interaction type as the preset interaction type; If not present, it is determined that the resource interaction type is not a preset interaction type.

[0026] Specifically, first, the resource flow record of the accessing user's application running the live broadcast program can be obtained, and then it is detected whether there is a resource interaction record in the resource flow record. If so, it indicates that the resource request submitted by the accessing user is not the first resource request submitted, that is, the accessing user has a historical resource interaction process before submitting this resource request, then the resource interaction type is determined to be the preset interaction type, or the resource interaction type can also be determined to be the first resource interaction type; if not, it indicates that the resource request submitted by the accessing user is the first resource request submitted, then the resource interaction type is determined not to be the preset interaction type, or, correspondingly, the resource interaction type can also be determined to be the second resource interaction type. In this case, when it is determined that the resource interaction type is not the preset interaction type, the preset resource value is returned to the user terminal.

[0027] It should be noted that, considering that the resource interaction records of the visiting user and other related data involved in this specification may belong to the privacy of the visiting user to a certain extent, if you want to collect or transmit the resource interaction records of the visiting user and other related data, you can obtain the authorization of the visiting user before collecting or transmitting the data, so that the operation of collecting or transmitting data complies with relevant data management regulations. For example, the visiting user can perform data authorization when starting the live broadcast program, and can also perform data authorization when the visiting user first accesses the live broadcast program; the specific method of data authorization can be to send a user data authorization reminder to the visiting user, and the visiting user can obtain data collection authorization after confirming the reminder through an instruction, or, the method of data authorization can also be to obtain data collection or data transmission authorization by signing a data authorization agreement.

[0028] Step S206: If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the intelligent agent calling the large language model to perform resource interaction recommendation.

[0029] During specific implementation, after determining the resource interaction type based on the resource interaction record of the visiting user, if the resource interaction type is a preset type, the resource interaction value obtained and stored by the intelligent agent calling the large language model for resource interaction recommendation is read; specifically, in order to provide the visiting user with a resource interaction value that is more in line with the visiting user's preferences in the subsequent resource processing process, and thereby improve the visiting user's satisfaction and retention rate in the subsequent resource processing process, the resource interaction value can be read after the intelligent agent calls the large language model for resource interaction recommendation to obtain and store the resource interaction value.

[0030] During the specific implementation process, the agent that calls the large language model can use the intelligent development platform to generate the agent. At the same time, in order to improve the relevance and accuracy of the subsequent resource interactive recommendation by the large language model, a prompt word template can be further generated on the intelligent development platform to guide the large language model to understand and process input data through the prompt word template. In an optional implementation provided by this embodiment, the agent is constructed in the following manner: Generate an intelligent agent based on the intelligent agent configuration information configured on the intelligent development platform; Generate a prompt word template for the large language model based on the prompt words uploaded to the intelligent development platform.

[0031] Among them, a large language model (LLM) refers to a pre-trained natural language model. The large language model can adopt a foundation model (Foundation Models) or a pre-trained model. The architecture of the large language model can be a neural network architecture, Transform architecture, or other architecture with large batch parameters. The specific large language model can directly adopt the foundation model or pre-trained model. It can also fine-tune the foundation model or pre-trained model based on the foundation model or pre-trained model for the specific task of resource interaction recommendation to obtain a large language model that can perform the specific task of resource interaction recommendation.

[0032] Optionally, the large language model includes a question-answering language model or a reasoning language model provided by the intelligent development platform.

[0033] Specifically, in the process of building an intelligent agent through the intelligent development platform, the creation of an initialized intelligent agent can be performed according to the configuration interface provided by the intelligent agent development platform. Based on the creation of the initialized intelligent agent, the intelligent agent can be configured according to the intelligent agent configuration information provided by the intelligent development platform, and the prompt word template of the large language model can be generated from the prompt words uploaded on the intelligent development platform.

[0034] Optionally, the prompt word template includes: an input rule field, an output rule field, a recommendation description field for recommending a role description and / or a recommended scenario description for making resource interaction recommendations, and / or a task description field and / or a rule description field for making interactive numerical recommendations for resource interactions to users accessing the live broadcast program.

[0035] For example, first, an agent is initialized through the creation interface provided by the intelligent development platform to obtain an initialized agent; second, basic configuration information of the initialized agent is configured, where the basic configuration information includes the agent's name, description, creation user, icon, and / or configuration mode; further, the agent is configured in mode, specifically, the agent can be configured in general agent mode, and further, the agent can be configured in multi-agent mode; further, prompt words are written, where the prompt words specifically refer to natural language prompt words that are subsequently input into the large language model, so that the large language model can perform corresponding processing based on the natural language prompt words; Specifically, when writing prompt words, you can write prompt words yourself, such as writing task descriptions, adding constraints and / or answer examples. You can also call the prompt word generation model according to the configured interface to generate prompt words, or you can generate prompt words using the prompt word template recommended by the intelligent development platform and modify the prompt words. For another example, in the process of configuring a large language model, the mode of the large language model can be configured according to the function of the large language model. For example, the large language model can be configured as a question-answering language model. In this case, the large language model can receive questions as input, locate and extract information by analyzing the relevant text of the question and / or information in the database to provide answers; or the large language model can be configured as a reasoning language model. In this case, after receiving the input, the large language model can understand the surface meaning of the language and perform deep logical inference based on the existing information to provide new conclusions and / or suggestions; in addition, in addition to configuring the mode of the large language model according to the function of the large language model, the mode of the large language model can also be configured according to the process of the large language model. For example, the large language model can be configured as an end-to-end mode large language model, or the large language model can be configured as a segmented mode large language model.

[0036] In practical applications, in order to make the prompt words subsequently input into the large language model more accurate, and thus make the recommendation results output by the large language model for interactive resource recommendation more in line with the needs of the accessing user, after building an intelligent agent through the intelligent development platform, the generated intelligent agent can also call the large language model for interactive resource recommendation to obtain debugging results and continuously adjust the prompt words, so that the prompt words subsequently input into the large language model are more accurate. An optional implementation provided by this embodiment also includes: Based on the debugging data input into the intelligent development platform, the generated intelligent agent calls the large language model to perform resource interactive recommendation to obtain the debugging results.

[0037] Specifically, on the basis of constructing an intelligent agent, debugging data can be input into the intelligent development platform to enable the intelligent agent to call the big language for resource interaction. Specifically, in the process of the intelligent agent calling the big language model for resource interaction, the intelligent agent can convert the debugging data into a data format suitable for processing by the big language model according to a pre-set prompt word template and submit it to the big language model. The big language model can make resource interaction recommendations based on the input debugging data and return the debugging results to the intelligent agent. Based on this, the debugging results of the resource interaction recommendation called by the generated intelligent agent can be obtained. The debugging results may include the recommendation results and basis for the resource interaction recommendation made by the intelligent agent according to the intelligent agent configuration information and / or prompt word template.

[0038] Furthermore, based on the debugging results obtained by the generated intelligent agent calling the large language model for resource interactive recommendation, the performance of the intelligent agent can be evaluated based on the debugging results to see whether it meets the expected goals. For example, if it is found that the debugging results deviate greatly from expectations, or the debugging results do not fully consider the changes in the preferences of the visiting users, the configuration information and / or prompt word template of the intelligent agent can be adjusted, such as modifying the parameters in the prompt word template and / or adjusting the configuration information of the intelligent agent; the process of obtaining the debugging results by the generated intelligent agent calling the large language model for resource interactive recommendation based on the debugging data input on the intelligent development platform and adjusting the configuration information and / or prompt word template of the intelligent agent based on the debugging results is continuously iterated until an optimized intelligent agent with stable performance is obtained. Based on this, an intelligent agent that calls the large language model for resource interactive recommendation can be obtained; optionally, the debugging results are used to adjust the configuration information or prompt word template of the generated intelligent agent.

[0039] In a specific implementation, in the process of interactively recommending resources by an agent calling a large language model, in order to improve the overall response speed and concurrent processing capability of the system during resource processing, the interactive resource recommendation processing task can be asynchronously performed to complete the interactive resource recommendation without delaying the user interface response of the access user. In an optional implementation provided by this embodiment, interactive resource recommendation by an agent calling a large language model includes: Create asynchronous tasks through the agent and submit them to the thread pool for task thread allocation; Execute asynchronous tasks on the assigned task thread through the agent.

[0040] Specifically, first, an asynchronous task can be created through an intelligent agent and submitted to a thread pool, which manages and allocates specific task threads for the asynchronous task. During the task thread allocation process, each asynchronous task can be allocated to an independent thread to run, so as to avoid overall performance degradation due to blocking of a single thread. Specifically, in the process of creating an asynchronous task through an intelligent agent, the creation of the asynchronous task is performed after the resource interaction of the previous resource request of the user is accessed is processed. During the execution of the assigned task thread, the asynchronous task can be executed through the intelligent agent, and specifically, the large language model can be called by the intelligent agent to execute the asynchronous task. Optionally, the asynchronous task is created after the resource interaction of the previous resource request of the user is accessed is processed. Optionally, the large language model is called during the execution of the asynchronous task.

[0041] In the process of executing the asynchronous task, in order to ensure the accuracy and relevance of the resource interaction recommendation, the resource interaction record of the accessing user can be used to make the resource interaction recommendation. In an optional implementation provided by this embodiment, the execution of the asynchronous task includes: Generate prompt text according to the prompt word template and the resource interaction value recorded in the read resource interaction record; The prompt text is input into the large language model for resource interaction recommendation to obtain the resource interaction value.

[0042] Specifically, during the process of executing asynchronous tasks in the assigned task thread, the intelligent agent can first obtain the resource interaction records of the accessing user, and then the intelligent agent can generate a prompt text based on the pre-set prompt word template and the resource interaction value recorded in the resource interaction record, and input the prompt text into the large language model, so that the large language model can make resource interaction recommendations based on the prompt text and output the resource interaction value. The large language model then returns the resource interaction value to the intelligent agent, and based on this, the resource interaction value is obtained.

[0043] Thereafter, after executing the asynchronous task and obtaining the resource interaction value, in order to facilitate the subsequent acquisition and use of the resource interaction value in the process of resource interaction recommendation, the resource interaction value can be stored in a preset storage area, for example, the resource interaction value can be stored in a database, and can also be stored in a cache; optionally, the resource interaction value obtained by executing the asynchronous task is stored in a preset storage area; the resource interaction value includes the resource interaction value stored in the preset storage area that is read.

[0044] Specifically, in the process of interactive resource recommendation, in order to ensure the accuracy of the recommendation results obtained by interactive resource recommendation, semantic recognition can be performed on the prompt text. At the same time, the prompt text is input into the processing network to perform the recommendation operation, which also improves processing efficiency. In an optional implementation provided by this embodiment, interactive resource recommendation includes: Perform semantic recognition on the recommendation description field, input rule field, task description field, and rule description field contained in the prompt text, and determine the recommended action based on the recognition result; The resource interaction value contained in the prompt text is input into the processing network to perform the recommendation operation to obtain the resource interaction value, and the recommendation result containing the resource interaction value is generated according to the semantic recognition result of the output rule field.

[0045] For example, during resource interaction recommendation, the prompt text includes the following recommendation description: "You are a resource interaction value recommendation assistant named 'Resource Interaction Recommendation Assistant'. Your goal is to improve the satisfaction and long-term retention rate of visiting users during resource interactions. At the same time, you provide the resource interaction value of a single resource interaction performed by the visiting user within a reasonable range, ensuring that the recommended resource interaction value results are explainable and logical." The input rule field is as follows: "When inputting, a list of resource interaction values is given in reverse chronological order of the most recent resource interaction record;"; The task description field is as follows: "Recommended resource interaction values are ranked according to the resource interaction values that are most likely to be completed by the visiting user, with this value having the highest weight. While ensuring visiting user satisfaction and retention rate, the recommended resource interaction value should be appropriately increased. The recommended resource interaction value should not exceed a% of the visiting user's historical highest single resource interaction value. The recommended resource interaction value cannot be repeated." The rule description field is as follows: "Output the recommendation results in JSON format. The recommendation results contain two keys: value and reason. The value corresponds to the recommended b resource interaction values, separated by commas. The reason corresponds to the analysis reason." Perform semantic recognition on the recommendation description field, input rule field, task description field, and rule description field contained in the prompt text, determine the recommended action based on the recognition result, and input the resource interaction values contained in the prompt text into the processing network to execute the recommended action; the resource interaction values are as follows: x1, x2, x2, x3, x2, x2, x2, x1, x4, x5, x4, x1, x1, x1, x1, x6, x1, x1, x1, x7; The recommended operation is performed based on the resource interaction value contained in the prompt text to obtain the resource interaction value, and the recommendation result containing the resource interaction value is generated according to the semantic recognition result of the output rule field as follows: "value": "x3, x5, x4, x6, x7"; "reason": "The user has recently frequently selected resource interaction values such as x3, x2, and x4. The highest resource interaction value for user interaction is x4. Therefore, the recommended resource interaction value should not exceed a% of x4, that is, no more than x4*a%. The recommended resource interaction value avoids duplication and is within the user's preferred range. At the same time, the resource interaction value is appropriately increased to improve user satisfaction and long-term retention." Step S208: Return the resource interaction value to the user terminal, and perform resource interaction processing according to the target resource value submitted by the user terminal.

[0046] During specific implementation, in order to improve the satisfaction and retention rate of visiting users, when the resource interaction value is obtained, the resource interaction value is returned to the user terminal, so that the user terminal can obtain and display the resource interaction value to the visiting user, and when the target resource value submitted by the user terminal is obtained, resource interaction processing is performed according to the target resource data; specifically, after receiving the returned resource interaction value and displaying it, the user terminal can obtain the target resource value selected by the visiting user in the resource interaction value and submit it. Based on this, resource interaction processing can be performed according to the target resource value submitted by the user terminal.

[0047] During the specific implementation process, in order to improve the efficiency and success rate of resource interaction processing and reduce the reduction in user retention rate caused by the long waiting time of the accessing user during the resource interaction processing, real-time resource interaction processing can be performed based on the target resource value selected by the accessing user. In an optional implementation provided by this embodiment, resource interaction processing is performed based on the target resource value submitted by the user terminal, including: Purchase virtual resources of the live broadcast program according to the target resource value, and transfer the purchased virtual resources to the program account of the accessing user in the live broadcast program.

[0048] Specifically, after the user terminal obtains and submits the target resource value selected by the accessing user in the resource interaction value, the virtual resource of the live broadcast program is purchased according to the target resource value submitted by the user terminal. Specifically, the payment resources for purchasing the virtual resources can be deducted from the application account of the accessing user's application running the live broadcast program, and the corresponding purchased virtual resources are transferred to the accessing user's program account in the live broadcast program. In addition, after the resource interaction processing is completed, the resource interaction records such as the resource interaction values of the resource interaction processing performed by the accessing user can also be retained and recorded for subsequent query during the resource processing process. Optionally, the payment resources for purchasing the virtual resources are deducted from the application account of the application running the live broadcast program.

[0049] In summary, the present embodiment provides one or more resource processing methods for live broadcast programs. During the resource processing process, after the user terminal generates and submits a resource request for resource interaction based on the resource interaction operation submitted by the visiting user in the live broadcast program, the method accordingly receives the resource request submitted by the user terminal; then, the resource interaction type of the resource request is determined based on the resource interaction record of the visiting user. When the resource interaction type is determined to be a preset interaction type, in order to provide the visiting user with a resource interaction value that is more in line with the visiting user's preferences in the subsequent resource processing process, thereby improving the visiting user's satisfaction and retention rate in the subsequent resource processing process, the intelligent agent can call a large language model to perform resource interaction recommendation to obtain and store the resource interaction value, and then read the resource interaction value; finally, when the resource interaction value is obtained, the resource interaction value is returned to the user terminal so that the user terminal obtains and displays the resource interaction value to the visiting user, and when the target resource value submitted by the user terminal is obtained, resource interaction processing is performed according to the target resource data, thereby improving the visiting user's experience and retention rate of resource interaction by reducing the visiting user's operation steps and saving the visiting user's operation time in the resource processing process.

[0050] Furthermore, in the process of calling the large language model through the intelligent agent to perform resource interactive recommendation, in order to improve the response speed and concurrent processing capability of the resource processing process, so as to complete the resource interactive recommendation without delaying the user interface response of the accessing user, the intelligent agent can create an asynchronous task, submit the asynchronous task to the thread pool for task thread allocation, and execute the asynchronous task through the intelligent agent in the allocated task thread; wherein, the intelligent agent can be constructed through the intelligent development platform, specifically, the intelligent agent can be generated according to the intelligent agent configuration information configured on the intelligent development platform, and the prompt word template of the large language model can be generated according to the prompt words uploaded on the intelligent development platform. At the same time, in order to make the prompt words subsequently input into the large language model more accurate, and thus make the recommendation results output by the large language model for resource interactive recommendation more in line with the needs of the accessing user, the debugging data input on the intelligent development platform can be used to obtain the debugging results by the generated intelligent agent calling the large language model for resource interactive recommendation, and the configuration information or prompt word template of the generated intelligent agent can be adjusted according to the debugging results, thereby further improving the accuracy of the recommendation results output by the large language model for resource interactive recommendation and reducing the probability of delayed response of the user interface of the accessing user, thereby further improving the satisfaction of the accessing user with resource interaction.

[0051] The above steps S202 to S208 provided in this embodiment can be executed by the server. It should be noted that the above steps S202 to S208 executed by the server and the steps S402 to S406 executed by the user terminal of the accessing user in the following embodiment can cooperate with each other during the execution process. Therefore, when reading this embodiment, please refer to the corresponding contents of steps S402 to S406 provided in the following method embodiment, and when reading the following method embodiment, please refer to the corresponding contents of steps S202 to S208 provided in this embodiment.

[0052] The following is an example of an application of a resource processing method provided in this embodiment in a resource processing scenario. Figure 3 , further describes the resource processing method of the live broadcast program provided in this embodiment, see Figure 3 , a resource processing method for a live broadcast program applied to a resource processing scenario specifically includes the following steps.

[0053] Step S306: receiving a resource request submitted by the user terminal for accessing resources for user interaction in the live broadcast program.

[0054] Step S308, detect whether there is a resource interaction record of the live broadcast program in the resource flow record of the application program of the accessing user running the live broadcast program; if so, execute step S310; if not, determine that the resource interaction type is not the preset interaction type, and return the preset resource value to the user terminal.

[0055] Step S310: Determine whether the resource interaction type is a preset interaction type.

[0056] Step S312: Create an asynchronous task through the agent and submit the asynchronous task to the thread pool for task thread allocation.

[0057] Optionally, the intelligent agent is constructed in the following manner: the intelligent agent is generated according to the intelligent agent configuration information configured on the intelligent development platform; and a prompt word template of the large language model is generated according to the prompt words uploaded on the intelligent development platform.

[0058] Step S314: The intelligent agent generates a prompt text according to the prompt word template and the resource interaction value recorded in the read resource interaction record.

[0059] Step S316: The prompt text is input into the large language model through the intelligent agent for interactive resource recommendation.

[0060] Optionally, the resource interaction recommendation includes: performing semantic recognition on the recommendation description field, input rule field, task description field and / or rule description field contained in the prompt text, and determining the recommended operation based on the recognition result; obtaining the resource interaction value by inputting the resource interaction value contained in the prompt text into the processing network to perform the recommendation operation, and generating a recommendation result containing the resource interaction value according to the semantic recognition result of the output rule field.

[0061] Step S318: Read the resource interaction value obtained and stored by performing resource interaction recommendation.

[0062] Step S320: Return the resource interaction value to the user terminal.

[0063] Step S328: Acquire the target resource value submitted by the user terminal.

[0064] Step S330: Purchase virtual resources of the live broadcast program according to the target resource value, and transfer the purchased virtual resources to the program account of the accessing user in the live broadcast program.

[0065] It should be noted that any one of steps S306 to S320 and any combination of multiple steps from steps S328 to S330 can be combined with any one of steps S202 to S208 to form a new implementation method according to the needs of implementation deployment; in addition, according to the needs of actual deployment, any one or multiple technical features can be selected from steps S306 to S320 and steps S328 to S330 and combined with any one or multiple technical features provided by steps S202 to S208 to form a new implementation method; or, any one or multiple technical features from steps S306 to S320 and steps S328 to S330 can be replaced with any one or multiple technical features provided by steps S202 to S208 to form a new implementation method according to the needs of actual deployment, which will not be repeated here.

[0066] In addition, it should be noted that the above-mentioned steps S306 to S320 and steps S328 to S330 provided in this embodiment can be executed by the server. It should be noted that the above-mentioned steps S306 to S320 and steps S328 to S330 executed by the server and steps S302 to S304 and steps S322 to S326 executed by the user terminal of the accessing user in the following embodiment can cooperate with each other during the execution process. Therefore, when reading this embodiment, please refer to the corresponding contents of steps S302 to S304 and steps S322 to S326 provided in the following method embodiment, and when reading the following method embodiment, please refer to the corresponding contents of steps S306 to S320 and steps S328 to S330 provided in this embodiment.

[0067] One or more embodiments of another method for processing resources of a live broadcast program provided in this specification are as follows: Reference Figure 4 The resource processing method of the live broadcast program provided in this embodiment specifically includes steps S402 to S406.

[0068] Step S402: Generate a resource request for resource interaction based on the resource interaction operation submitted by the accessing user in the live broadcast program and submit it to the server.

[0069] The live broadcast program described in this embodiment refers to a program that can be used for live broadcasting. The live broadcast program can specifically be a live broadcast subroutine, or can also be a service subroutine for live broadcasting in a resource service program. The live broadcast program can run in an application program for live broadcasting.

[0070] In this embodiment, the visiting user can interact with the host who is performing the live broadcast while accessing the live broadcast program. For example, the visiting user can donate resources to the host, or the visiting user can use resources to redeem virtual objects and donate the redeemed virtual objects to the host. Among them, in the process of the visiting user donating resources and / or redeeming resources to the host, in order to enable the visiting user to have sufficient resources for interaction, the visiting user can obtain resources by submitting resource interaction operations to perform resource interaction.

[0071] Among them, resource interaction refers to the interactive processing of resources by visiting users, specifically refers to the purchase or recharge of resources by visiting users in the live broadcast program; among them, resources refer to virtual resources, specifically refers to virtual resources that can be purchased, recharged, donated to the anchor and / or used in the live broadcast program.

[0072] In specific implementation, when an accessing user accesses online live broadcast through a live broadcast program, the accessing user can submit resource interaction operations in the live broadcast program. Accordingly, a resource request for resource interaction can be generated based on the resource interaction operation submitted by the accessing user and submitted to the server, so that the server obtains the resource request and performs subsequent resource processing based on the resource request.

[0073] Specifically, during the process of submitting resource interaction operations to the live broadcast program, the accessing user can submit a trigger operation through the resource interaction interface of the live broadcast program. Accordingly, the resource interaction operation includes a trigger operation for the resource interaction interface of the live broadcast program. In this case, based on the resource interaction operation submitted by the accessing user in the live broadcast program, a resource request for resource interaction is generated and submitted to the server. This can be replaced by: based on the trigger operation of the accessing user for the resource interaction interface of the live broadcast program, a resource request for resource interaction is generated and submitted to the server.

[0074] Step S404: receiving and displaying the resource interaction value returned by the server.

[0075] As described above, after a resource request is generated and submitted to the server, the server may receive the resource request. When the server receives the resource request, in order to improve the satisfaction and retention rate of the accessing user in the subsequent resource processing process, and thereby promote the success rate of resource interaction, the server may first determine the resource interaction type of the resource request based on the resource interaction record of the accessing user.

[0076] Among them, resource interaction records refer to records of resource interaction operations performed by accessing users in live broadcast programs. For example, resource interaction records may include the number of resource interactions and / or resource interaction time of accessing users. In addition, resource interaction records may also include other data related to resource interaction operations, such as resource interaction methods; the resource interaction type refers to the type of resource interaction operation performed by accessing users. For example, the resource interaction type may be the resource interaction type of the accessing user performing the resource interaction operation for the first time, or it may be the resource interaction type of the accessing user performing the resource interaction operation for a non-first time.

[0077] Specifically, in the process of determining the resource interaction type of the resource request based on the resource interaction record of the accessing user, the server may determine whether the resource interaction type is the resource interaction type of the first resource interaction operation by detecting whether there is a resource interaction record of the accessing user, thereby returning different resource values to the accessing user in the subsequent resource processing process for different resource interaction types. In an optional implementation manner provided in this embodiment, the server performs the following operations: Detect whether there is a resource interaction record of the live broadcast program in the resource flow record of the application program of the accessing user running the live broadcast program; If it exists, determine the resource interaction type as the preset interaction type; If not present, it is determined that the resource interaction type is not the preset interaction type.

[0078] Specifically, first, the server can obtain the resource flow record of the accessing user's application running the live broadcast program, and then detect whether there is a resource interaction record in the resource flow record. If so, it indicates that the resource request submitted by the accessing user is not the first resource request submitted, that is, the accessing user has a historical resource interaction process before submitting this resource request. The resource interaction type can be determined to be the preset interaction type, or the resource interaction type can also be determined to be the first resource interaction type; if not, it indicates that the resource request submitted by the accessing user is the first resource request submitted, and the resource interaction type can be determined not to be the preset interaction type, or, correspondingly, the resource interaction type can also be determined to be the second resource interaction type.

[0079] Furthermore, after the server determines the resource interaction type based on the resource interaction record of the accessing user, if the resource interaction type is not the preset interaction type, the server returns the preset resource value. Accordingly, the preset resource value returned by the server is received, and the resource interaction value is displayed on the resource interaction interface displayed after the above-mentioned resource interaction interface is triggered; conversely, if the resource interaction type is the preset type, the server reads the resource interaction value obtained and stored by the intelligent agent calling the large language model for resource interaction recommendation.

[0080] Among them, when the server determines that the resource interaction type is the preset interaction type, in order to provide the visiting user with a resource interaction value that is more in line with the visiting user's preferences in the subsequent resource processing process, and thereby improve the visiting user's satisfaction and retention rate in the subsequent resource processing process, the server can read the resource interaction value after the intelligent body calls the large language model to perform resource interaction recommendation to obtain and store the resource interaction value.

[0081] During the specific implementation process, the agent that calls the large language model can use the intelligent development platform to generate the agent. At the same time, in order to improve the relevance and accuracy of the subsequent resource interactive recommendation by the large language model, a prompt word template can be further generated on the intelligent development platform to guide the large language model to understand and process input data through the prompt word template. In an optional implementation provided by this embodiment, the agent is constructed in the following manner: Generate an intelligent agent based on the intelligent agent configuration information configured on the intelligent development platform; Generate a prompt word template for the large language model based on the prompt words uploaded to the intelligent development platform.

[0082] Among them, a large language model (LLM) refers to a pre-trained natural language model. The large language model can adopt a foundation model (Foundation Models) or a pre-trained model. The architecture of the large language model can be a neural network architecture, Transform architecture, or other architecture with large batch parameters. The specific large language model can directly adopt the foundation model or pre-trained model. It can also fine-tune the foundation model or pre-trained model based on the foundation model or pre-trained model for the specific task of resource interaction recommendation to obtain a large language model that can perform the specific task of resource interaction recommendation.

[0083] Optionally, the large language model includes a question-answering language model or a reasoning language model provided by the intelligent development platform.

[0084] Specifically, in the process of building an intelligent agent through the intelligent development platform, the creation of an initialized intelligent agent can be performed according to the configuration interface provided by the intelligent agent development platform. Based on the creation of the initialized intelligent agent, the intelligent agent can be configured according to the intelligent agent configuration information provided by the intelligent development platform, and the prompt word template of the large language model can be generated from the prompt words uploaded on the intelligent development platform.

[0085] Optionally, the prompt word template includes: an input rule field, an output rule field, a recommendation description field for recommending a role description and / or a recommended scenario description for making resource interaction recommendations, and / or a task description field and / or a rule description field for making interactive numerical recommendations for resource interactions to users accessing the live broadcast program.

[0086] For example, first, an agent is initialized through the creation interface provided by the intelligent development platform to obtain an initialized agent; second, basic configuration information of the initialized agent is configured, where the basic configuration information includes the agent's name, description, creation user, icon, and / or configuration mode; further, the agent is configured in mode, specifically, the agent can be configured in general agent mode, and further, the agent can be configured in multi-agent mode; further, prompt words are written, where the prompt words specifically refer to natural language prompt words that are subsequently input into the large language model, so that the large language model can perform corresponding processing based on the natural language prompt words; Specifically, when writing prompt words, you can write prompt words yourself, such as writing task descriptions, adding constraints and / or answer examples. You can also call the prompt word generation model according to the configured interface to generate prompt words, or you can generate prompt words using the prompt word template recommended by the intelligent development platform and modify the prompt words. For another example, in the process of configuring a large language model, the mode of the large language model can be configured according to the function of the large language model. For example, the large language model can be configured as a question-answering language model. In this case, the large language model can receive questions as input, locate and extract information by analyzing the relevant text of the question and / or information in the database to provide answers; or the large language model can be configured as a reasoning language model. In this case, after receiving the input, the large language model can understand the surface meaning of the language and perform deep logical inference based on the existing information to provide new conclusions and / or suggestions; in addition, in addition to configuring the mode of the large language model according to the function of the large language model, the mode of the large language model can also be configured according to the process of the large language model. For example, the large language model can be configured as an end-to-end mode large language model, or the large language model can be configured as a segmented mode large language model.

[0087] In practical applications, in order to make the prompt words subsequently input into the large language model more accurate, and thus make the recommendation results output by the large language model for interactive resource recommendation more in line with the needs of the accessing user, after building an intelligent agent through the intelligent development platform, the generated intelligent agent can also call the large language model for interactive resource recommendation to obtain debugging results and continuously adjust the prompt words, so that the prompt words subsequently input into the large language model are more accurate. An optional implementation provided by this embodiment also includes: Based on the debugging data input into the intelligent development platform, the generated intelligent agent calls the large language model to perform resource interactive recommendation to obtain the debugging results.

[0088] Specifically, on the basis of constructing an intelligent agent, debugging data can be input into the intelligent development platform to enable the intelligent agent to call the big language for resource interaction. Specifically, in the process of the intelligent agent calling the big language model for resource interaction, the intelligent agent can convert the debugging data into a data format suitable for processing by the big language model according to a pre-set prompt word template and submit it to the big language model. The big language model can make resource interaction recommendations based on the input debugging data and return the debugging results to the intelligent agent. Based on this, the debugging results of the resource interaction recommendation called by the generated intelligent agent can be obtained. The debugging results may include the recommendation results and basis for the resource interaction recommendation made by the intelligent agent according to the intelligent agent configuration information and / or prompt word template.

[0089] Furthermore, based on the debugging results obtained by the generated intelligent agent calling the large language model for resource interactive recommendation, the performance of the intelligent agent can be evaluated based on the debugging results to see whether it meets the expected goals. For example, if it is found that the debugging results deviate greatly from expectations, or the debugging results do not fully consider the changes in the preferences of the visiting users, the configuration information and / or prompt word template of the intelligent agent can be adjusted, such as modifying the parameters in the prompt word template and / or adjusting the configuration information of the intelligent agent; the process of obtaining the debugging results by the generated intelligent agent calling the large language model for resource interactive recommendation based on the debugging data input on the intelligent development platform and adjusting the configuration information and / or prompt word template of the intelligent agent based on the debugging results is continuously iterated until an optimized intelligent agent with stable performance is obtained. Based on this, an intelligent agent that calls the large language model for resource interactive recommendation can be obtained; optionally, the debugging results are used to adjust the configuration information or prompt word template of the generated intelligent agent.

[0090] In a specific implementation, in the process of interactive resource recommendation by an intelligent agent calling a large language model, in order to improve the overall response speed and concurrent processing capability of the system during resource processing, the interactive resource recommendation processing task can be asynchronously performed to complete the interactive resource recommendation without delaying the user interface response of the access user. In an optional implementation provided by this embodiment, the intelligent agent calls a large language model to perform interactive resource recommendation, including: Create asynchronous tasks through the agent and submit them to the thread pool for task thread allocation; Execute asynchronous tasks on the assigned task thread through the agent.

[0091] Specifically, first, an asynchronous task can be created through an intelligent agent and submitted to a thread pool, which manages and allocates specific task threads for the asynchronous task. During the task thread allocation process, each asynchronous task can be allocated to an independent thread to run, so as to avoid overall performance degradation due to blocking of a single thread. Specifically, in the process of creating an asynchronous task through an intelligent agent, the creation of the asynchronous task is performed after the resource interaction of the previous resource request of the user is accessed is processed. During the execution of the assigned task thread, the asynchronous task can be executed through the intelligent agent, and specifically, the large language model can be called by the intelligent agent to execute the asynchronous task. Optionally, the asynchronous task is created after the resource interaction of the previous resource request of the user is accessed is processed. Optionally, the large language model is called during the execution of the asynchronous task.

[0092] In the process of executing the asynchronous task, in order to ensure the accuracy and relevance of the resource interaction recommendation, the resource interaction record of the accessing user can be used to make the resource interaction recommendation. In an optional implementation provided by this embodiment, the execution of the asynchronous task includes: Generate prompt text according to the prompt word template and the resource interaction value recorded in the read resource interaction record; The prompt text is input into the large language model for resource interaction recommendation to obtain the resource interaction value.

[0093] Specifically, during the process of executing asynchronous tasks in the assigned task thread, the intelligent agent can first obtain the resource interaction records of the accessing user, and then the intelligent agent can generate a prompt text based on the pre-set prompt word template and the resource interaction value recorded in the resource interaction record, and input the prompt text into the large language model, so that the large language model can make resource interaction recommendations based on the prompt text and output the resource interaction value. The large language model then returns the resource interaction value to the intelligent agent, and based on this, the resource interaction value is obtained.

[0094] Thereafter, after executing the asynchronous task and obtaining the resource interaction value, in order to facilitate the subsequent acquisition and use of the resource interaction value in the process of resource interaction recommendation, the resource interaction value can be stored in a preset storage area, for example, the resource interaction value can be stored in a database, and can also be stored in a cache; optionally, the resource interaction value obtained by executing the asynchronous task is stored in a preset storage area; the resource interaction value includes the resource interaction value stored in the preset storage area that is read.

[0095] Specifically, in the process of interactive resource recommendation, in order to ensure the accuracy of the recommendation results obtained by interactive resource recommendation, semantic recognition can be performed on the prompt text. At the same time, the prompt text is input into the processing network to perform the recommendation operation, which also improves processing efficiency. In an optional implementation provided by this embodiment, interactive resource recommendation includes: Perform semantic recognition on the recommendation description field, input rule field, task description field, and rule description field contained in the prompt text, and determine the recommended action based on the recognition result; The resource interaction value contained in the prompt text is input into the processing network to perform the recommendation operation to obtain the resource interaction value, and the recommendation result containing the resource interaction value is generated according to the semantic recognition result of the output rule field.

[0096] For example, during resource interaction recommendation, the prompt text includes the following recommendation description: "You are a resource interaction value recommendation assistant named 'Resource Interaction Recommendation Assistant'. Your goal is to improve the satisfaction and long-term retention rate of visiting users during resource interactions. At the same time, you provide the resource interaction value of a single resource interaction performed by the visiting user within a reasonable range, ensuring that the recommended resource interaction value results are explainable and logical." The input rule field is as follows: "When inputting, a list of resource interaction values is given in reverse chronological order of the most recent resource interaction record;"; The task description field is as follows: "Recommended resource interaction values are ranked according to the resource interaction values that are most likely to be completed by the visiting user, with this value having the highest weight. While ensuring visiting user satisfaction and retention rate, the recommended resource interaction value should be appropriately increased. The recommended resource interaction value should not exceed a% of the visiting user's historical highest single resource interaction value. The recommended resource interaction value cannot be repeated." The rule description field is as follows: "Output the recommendation results in JSON format. The recommendation results contain two keys: value and reason. The value corresponds to the recommended b resource interaction values, separated by commas. The reason corresponds to the analysis reason." Perform semantic recognition on the recommendation description field, input rule field, task description field, and rule description field contained in the prompt text, determine the recommended action based on the recognition result, and input the resource interaction values contained in the prompt text into the processing network to execute the recommended action; the resource interaction values are as follows: x1, x2, x2, x3, x2, x2, x2, x1, x4, x5, x4, x1, x1, x1, x1, x6, x1, x1, x1, x7; The recommended operation is performed based on the resource interaction value contained in the prompt text to obtain the resource interaction value, and the recommendation result containing the resource interaction value is generated according to the semantic recognition result of the output rule field as follows: "value": "x3, x5, x4, x6, x7"; "reason": "The user has recently frequently selected resource interaction values such as x3, x2, and x4. The highest resource interaction value for user interaction is x4. Therefore, the recommended resource interaction value should not exceed a% of x4, that is, no more than x4*a%. The recommended resource interaction value avoids duplication and is within the user's preferred range. At the same time, the resource interaction value is appropriately increased to improve user satisfaction and long-term retention." In specific implementation, in order to improve the satisfaction and retention rate of visiting users, the server can return the resource interaction value when it obtains the resource interaction value. Correspondingly, the resource interaction value returned by the server is received and displayed; optionally, the resource interaction value is obtained by the intelligent agent calling the large language model for resource interaction recommendation.

[0097] Among them, in the process of receiving and displaying the resource interaction value for resource interaction returned by the server, in the case where a resource request is generated and submitted to the server based on the resource interaction operation of the accessing user in the resource interaction interface of the live broadcast program, the resource interaction value returned by the server can be displayed on the resource interaction interface displayed after the resource interaction interface is triggered. In an optional implementation manner provided by this embodiment, the resource interaction value for resource interaction returned by the server is received and displayed, including: receiving the resource interaction value for resource interaction returned by the server, and displaying the resource interaction value on the resource interaction interface displayed after the resource interaction interface is triggered.

[0098] Step S406: obtaining the target resource value selected by the accessing user from the resource interaction value and submitting it to the server so as to perform resource interaction processing according to the target resource value.

[0099] In specific implementation, after the resource interaction value for resource interaction returned by the above-mentioned receiving server is displayed, the visiting user can select the target resource value based on the displayed resource interaction value. Accordingly, the target resource value selected by the visiting user in the resource interaction value is obtained, and the target resource value is submitted to the server so that the server performs resource interaction processing based on the target resource value.

[0100] During the specific implementation process, in order to improve the efficiency and success rate of resource interaction processing and reduce the reduction in user retention rate caused by the long waiting time of the accessing user during the resource interaction processing, real-time resource interaction processing can be performed according to the target resource value selected by the accessing user. In an optional implementation provided by this embodiment, the server performs resource interaction processing according to the target resource value, including: Purchase virtual resources of the live broadcast program according to the target resource value, and transfer the purchased virtual resources to the program account of the accessing user in the live broadcast program.

[0101] Specifically, after obtaining the target resource value selected by the accessing user in the resource interaction value and submitting it to the server, the server obtains the target resource value and purchases the virtual resources of the live broadcast program according to the target resource value. The specific server can deduct the payment resources for purchasing the virtual resources from the application account of the accessing user's application running the live broadcast program, and transfer the corresponding purchased virtual resources to the program account of the accessing user in the live broadcast program; in addition, after the resource interaction processing is completed, the server can also retain and record the resource interaction records such as the resource interaction value of the resource interaction processing performed by the accessing user, so as to facilitate subsequent inquiries during the resource processing process. Optionally, the payment resources for purchasing the virtual resources are deducted from the application account of the application running the live broadcast program.

[0102] In summary, the present embodiment provides one or more resource processing methods for live broadcast programs. In the process of resource processing, a resource request for resource interaction is generated and submitted to the server based on the resource interaction operation submitted by the visiting user in the live broadcast program. When the server receives the resource request, in order to improve the satisfaction and retention rate of the visiting user in the subsequent resource processing process, and thus promote the success rate of resource interaction, the resource interaction type of the resource request can be determined based on the resource interaction record of the visiting user. When it is determined that the resource interaction type is a preset interaction type, the server can read the resource interaction value after the intelligent body calls the large language model to perform resource interaction recommendation to obtain and store the resource interaction value, so as to provide the visiting user with the resource interaction value in the subsequent resource processing process. A resource interaction value that is more in line with the preferences of the visiting user; then the server can return the obtained resource interaction value to the user terminal, and accordingly, the resource interaction value for resource interaction returned by the server is received and displayed; thereafter, after receiving the resource interaction value for resource interaction returned by the server and displaying it, the visiting user can select the target resource value according to the displayed resource interaction value, and accordingly, obtain the target resource value selected by the visiting user in the resource interaction value, and submit the target resource value to the server, so that the server performs resource interaction processing according to the target resource value, thereby improving the visiting user's experience and retention rate in resource interaction by reducing the visiting user's operation steps and saving the visiting user's operation time during the resource processing process.

[0103] The following takes the application of a resource processing method provided in this embodiment in a resource processing scenario as an example, combined with Figure 3 , further describes the resource processing method of the live broadcast program provided in this embodiment, see Figure 3 , a resource processing method for a live broadcast program applied to a resource processing scenario specifically includes the following steps.

[0104] Step S302: Generate a resource request for resource interaction based on the resource interaction operation submitted by the accessing user in the live broadcast program.

[0105] Step S304: Submit a resource request to the server.

[0106] Step S322: Receive and display the resource interaction value returned by the server.

[0107] Step S324: Obtain the target resource value selected by the accessing user in the resource interaction value.

[0108] Step S326: Submit the target resource value selected by the access user to the server.

[0109] It should be noted that any one of steps S302 to S304 and any combination of multiple steps in steps S322 to S326 can be combined with any one of steps S402 to S406 to form a new implementation method according to the needs of implementation deployment; in addition, according to the needs of actual deployment, any one or multiple technical features can be selected from steps S302 to S304 and steps S322 to S326 and combined with any one or multiple technical features provided in steps S402 to S406 to form a new implementation method; or, any one or multiple technical features in steps S302 to S304 and steps S322 to S326 can be replaced with any one or multiple technical features provided in steps S402 to S406 to form a new implementation method according to the needs of actual deployment, which will not be repeated here.

[0110] An embodiment of a resource processing device for a live broadcast program provided in this specification is as follows: In the above embodiment, a resource processing method for a live broadcast program is provided. Correspondingly, a resource processing device for a live broadcast program is also provided, which will be described below with reference to the accompanying drawings.

[0111] Reference Figure 5 , which shows a schematic diagram of an embodiment of a resource processing device for a live broadcast program provided by this embodiment.

[0112] Since the device embodiment corresponds to the method embodiment, the description is relatively simple. For the relevant parts, please refer to the corresponding description of the method embodiment provided above. The device embodiment described below is only illustrative.

[0113] This embodiment provides a resource processing device for a live broadcast program, the device comprising: The request receiving module 502 is configured to receive a resource request submitted by a user terminal for accessing a resource interaction performed by the user in the live broadcast program; An interaction type determination module 504 is configured to determine a resource interaction type of the resource request according to the resource interaction record of the accessing user; The value reading module 506 is configured to read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation if the resource interaction type is a preset interaction type; The resource interaction module 508 is configured to return the resource interaction value to the user terminal and perform resource interaction processing according to the target resource value submitted by the user terminal.

[0114] Another embodiment of a resource processing device for a live broadcast program provided in this specification is as follows: In the above embodiment, another method for processing resources of a live broadcast program is provided. Correspondingly, another device for processing resources of a live broadcast program is also provided, which will be described below with reference to the accompanying drawings.

[0115] Reference Figure 6 , which shows a schematic diagram of another embodiment of a resource processing device for a live broadcast program provided by this embodiment.

[0116] Since the device embodiment corresponds to the method embodiment, the description is relatively simple. For the relevant parts, please refer to the corresponding description of the method embodiment provided above. The device embodiment described below is only illustrative.

[0117] This embodiment provides a resource processing device for a live broadcast program, the device comprising: The request submission module 602 is configured to generate a resource request for resource interaction according to the resource interaction operation submitted by the accessing user in the live broadcast program and submit it to the server; The value display module 604 is configured to receive and display the resource interaction value returned by the server; the resource interaction value is obtained by the agent calling the large language model to perform resource interaction recommendation; The target value submission module 606 is configured to obtain the target resource value selected by the accessing user from the resource interaction values and submit it to the server so as to perform resource interaction processing according to the target resource value.

[0118] A server embodiment provided in this specification is as follows: Corresponding to the resource processing method of a live broadcast program described above, based on the same technical concept, one or more embodiments of this specification further provide a server, which is used to execute the resource processing method of a live broadcast program provided above. Figure 7 A schematic diagram of the structure of a server provided in one or more embodiments of this specification.

[0119] This embodiment provides a server, including: like Figure 7As shown, servers can vary significantly depending on their configuration or performance. They may include one or more processors 701 and memory 702. Memory 702 may store one or more applications or data. Memory 702 may be either ephemeral or persistent. Applications stored in memory 702 may include one or more modules (not shown), each of which may include a series of computer-executable instructions within the server. Furthermore, processor 701 may be configured to communicate with memory 702 to execute the series of computer-executable instructions within memory 702 on the server. The server may also include one or more power supplies 703, one or more wired or wireless network interfaces 704, one or more input / output interfaces 705, and the like.

[0120] In a specific embodiment, the server includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions in the server, and the one or more programs are configured to be executed by one or more processors, including computer-executable instructions for performing the following: Receive a resource request submitted by a user terminal for accessing resources for interaction with the user in the live broadcast program; Determining the resource interaction type of the resource request according to the resource interaction record of the accessing user; If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction value is returned to the user terminal, and resource interaction processing is performed according to the target resource value submitted by the user terminal.

[0121] An embodiment of a user terminal provided in this specification is as follows: Corresponding to the resource processing method of another live broadcast program described above, based on the same technical concept, one or more embodiments of this specification further provide a user terminal, which is used to execute the resource processing method of another live broadcast program provided above. Figure 8 A schematic diagram of the structure of a user terminal provided in one or more embodiments of this specification.

[0122] This embodiment provides a user terminal, including: like Figure 8As shown, user terminals can vary significantly due to configuration or performance differences and may include one or more processors 801 and memory 802. Memory 802 may store one or more applications or data. Memory 802 may be either transient or persistent. Applications stored in memory 802 may include one or more modules (not shown), each of which may comprise a series of computer-executable instructions within the user terminal. Furthermore, processor 801 may be configured to communicate with memory 802 to execute the series of computer-executable instructions within memory 802 on the user terminal. The user terminal may also include one or more power supplies 803, one or more wired or wireless network interfaces 804, one or more input / output interfaces 805, one or more keyboards 806, and the like.

[0123] In a specific embodiment, the user terminal includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer-executable instructions in the user terminal, and the one or more programs are configured to be executed by one or more processors, including computer-executable instructions for performing the following: Generate resource requests for resource interaction based on the resource interaction operations submitted by the accessing user in the live broadcast program and submit them to the server; Receive and display the resource interaction value returned by the server for resource interaction; the resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target resource value selected by the accessing user from the resource interaction value is obtained and submitted to the server, so as to perform resource interaction processing according to the target resource value.

[0124] An embodiment of a computer-readable storage medium provided in this specification is as follows: Corresponding to the resource processing method of a live broadcast program described above, based on the same technical concept, one or more embodiments of this specification also provide a computer-readable storage medium.

[0125] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions. When the computer-executable instructions are executed, the following process is implemented: Receive a resource request submitted by a user terminal for accessing resources for interaction with the user in the live broadcast program; Determining the resource interaction type of the resource request according to the resource interaction record of the accessing user; If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction value is returned to the user terminal, and resource interaction processing is performed according to the target resource value submitted by the user terminal.

[0126] It should be noted that the embodiment of a computer-readable storage medium in this specification and the embodiment of a resource processing method for a live broadcast program in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0127] Another embodiment of a computer-readable storage medium provided in this specification is as follows: Corresponding to the resource processing method of another live broadcast program described above, based on the same technical concept, one or more embodiments of this specification also provide another computer-readable storage medium.

[0128] The computer-readable storage medium provided in this embodiment is used to store computer-executable instructions. When the computer-executable instructions are executed, the following process is implemented: Generate resource requests for resource interaction based on the resource interaction operations submitted by the accessing user in the live broadcast program and submit them to the server; Receive and display the resource interaction value returned by the server for resource interaction; the resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target resource value selected by the accessing user from the resource interaction value is obtained and submitted to the server, so as to perform resource interaction processing according to the target resource value.

[0129] It should be noted that the embodiment of another computer-readable storage medium in this specification and the embodiment of another resource processing method of a live broadcast program in this specification are based on the same inventive concept. Therefore, the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0130] An embodiment of a computer program product provided in this specification is as follows: Corresponding to the resource processing method of a live broadcast program described above, based on the same technical concept, one or more embodiments of this specification also provide a computer program product.

[0131] A computer program product comprising a computer program / instructions, which, when executed by a processor, implements the following steps: Receive a resource request submitted by a user terminal for accessing resources for interaction with the user in the live broadcast program; Determining the resource interaction type of the resource request according to the resource interaction record of the accessing user; If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction value is returned to the user terminal, and resource interaction processing is performed according to the target resource value submitted by the user terminal.

[0132] It should be noted that an embodiment of a computer program product in this specification and an embodiment of a resource processing method in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0133] Another computer program product embodiment provided in this specification is as follows: Corresponding to the resource processing method of another live broadcast program described above, based on the same technical concept, one or more embodiments of this specification also provide another computer program product.

[0134] A computer program product comprising a computer program / instructions, which, when executed by a processor, implements the following steps: Generate resource requests for resource interaction based on the resource interaction operations submitted by the accessing user in the live broadcast program and submit them to the server; Receive and display the resource interaction value returned by the server for resource interaction; the resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target resource value selected by the accessing user from the resource interaction value is obtained and submitted to the server, so as to perform resource interaction processing according to the target resource value.

[0135] It should be noted that an embodiment of a computer program product in this specification and an embodiment of a resource processing method in this specification are based on the same inventive concept, so the specific implementation of this embodiment can refer to the implementation of the aforementioned corresponding method, and the repeated parts will not be repeated.

[0136] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments. For example, the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments are similar to the method embodiments, so the description is relatively simple. For relevant content in the device embodiments, equipment embodiments, computer-readable storage medium embodiments, and computer program product embodiments, please refer to the partial description of the method embodiments.

[0137] The foregoing description of this specification describes specific embodiments. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in an order different from that described in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order shown or the sequential order to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0138] In the 1930s, technological improvements could be clearly distinguished as either hardware improvements (for example, improvements to circuit structures like diodes, transistors, and switches) or software improvements (improvements to process flows). However, with the advancement of technology, many process flow improvements today can be considered direct improvements to hardware circuit structures. Designers almost always create the corresponding hardware circuit structure by programming the improved process flow into the hardware circuit. Therefore, it cannot be said that a process flow improvement cannot be implemented using physical hardware modules. For example, a programmable logic device (PLD), such as a field programmable gate array (FPGA), is an integrated circuit whose logical function is determined by user programming. Designers can "integrate" a digital system on a PLD by programming it themselves, without having to hire a chip manufacturer to design and manufacture a dedicated integrated circuit chip. Moreover, nowadays, instead of manually fabricating integrated circuit chips, this programming is mostly performed using software called a "logic compiler." This is similar to the software compilers used during program development. Before compilation, the original code must be written in a specific programming language, called a Hardware Description Language (HDL). There are many types of HDL, including ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, and RHDL (Ruby Hardware Description Language). Currently, the most commonly used are VHDL (Very-High-Speed Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art will also understand that simply by programming a method flow in one of these hardware description languages and then programming it into an integrated circuit, a hardware circuit that implements the logic method flow can be easily obtained.

[0139] The controller can be implemented in any suitable manner. For example, the controller can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, an application-specific integrated circuit (ASIC), a programmable logic controller, and an embedded microcontroller. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicone Labs C8051F320. The memory controller can also be implemented as part of the memory control logic. Those skilled in the art will also appreciate that, in addition to implementing the controller purely in computer-readable program code, the controller can also be implemented in the form of logic gates, switches, an application-specific integrated circuit, a programmable logic controller, an embedded microcontroller, etc. by logically programming the method steps. Therefore, such a controller can be considered a hardware component, and the means for implementing the various functions included therein can also be considered as structures within the hardware component. Alternatively, the means for implementing the various functions can be considered both a software module implementing the method and a structure within the hardware component.

[0140] The systems, devices, modules, or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a computer. Specifically, the computer may be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0141] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing the embodiments of this specification, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0142] Those skilled in the art will appreciate that one or more embodiments of this specification may be provided as a method, system, or computer program product. Thus, one or more embodiments of this specification may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, this specification may take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0143] This specification is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of this specification. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable resource processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable resource processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0144] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable resource processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0145] These computer program instructions can also be loaded onto a computer or other programmable resource processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0146] In a typical configuration, a computing device includes one or more processors (CPUs), input / output interfaces, network interfaces, and memory.

[0147] Memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0148] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented using any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer-readable storage media include, but are not limited to, phase-change RAM (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, magnetic disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory computer-readable media such as modulated data signals and carrier waves.

[0149] It should also be noted that the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, commodity, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, commodity, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising at least one ..." does not exclude the presence of additional identical elements in the process, method, commodity, or apparatus comprising the element.

[0150] One or more embodiments of this specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. One or more embodiments of this specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0151] The foregoing description is merely an example of the present invention and is not intended to limit the present invention. Persons skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be included within the scope of the claims herein.

Claims

1. A resource processing method for a live broadcast program, comprising: Receive a resource request submitted by a user terminal for accessing resources for interaction with the user in the live broadcast program; Determining the resource interaction type of the resource request according to the resource interaction record of the accessing user; If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction value is returned to the user terminal, and resource interaction processing is performed according to the target resource value submitted by the user terminal.

2. The resource processing method for a live broadcast program according to claim 1, wherein the method of performing resource interactive recommendation by calling a large language model through an intelligent agent comprises: Creating an asynchronous task through the agent and submitting the asynchronous task to the thread pool for task thread allocation; The asynchronous task is executed by the agent in the assigned task thread; the large language model is called during the execution of the asynchronous task.

3. The resource processing method for a live broadcast program according to claim 2, wherein the execution of the asynchronous task comprises: generating a prompt text according to a prompt word template and the resource interaction value recorded in the read resource interaction record; The prompt text is input into the large language model for resource interaction recommendation to obtain the resource interaction value.

4. The resource processing method for a live broadcast program according to claim 2, wherein the asynchronous task is created after the resource interaction processing of the previous resource request of the accessing user is completed; The resource interaction value obtained by executing the asynchronous task is stored in a preset storage area; the resource interaction value includes the resource interaction value stored in the preset storage area.

5. The resource processing method for a live broadcast program according to claim 3, wherein the prompt word template includes at least one of the following: Input rule field, output rule field, recommendation description field for recommended role description and / or recommended scenario description for resource interaction recommendation, task description field and / or rule description field for interactive numerical recommendation of resource interaction to users visiting the live broadcast program.

6. The resource processing method for a live broadcast program according to claim 5, wherein the resource interactive recommendation comprises: Performing semantic recognition on the recommendation description field, input rule field, task description field, and rule description field contained in the prompt text, and determining a recommended action based on the recognition result; The resource interaction value included in the prompt text is input into a processing network to perform a recommendation operation to obtain the resource interaction value, and a recommendation result including the resource interaction value is generated according to a semantic recognition result of an output rule field.

7. The resource processing method for a live broadcast program according to claim 1, wherein the agent is constructed in the following manner: Generate an intelligent agent based on the intelligent agent configuration information configured on the intelligent development platform; Generate a prompt word template of the large language model according to the prompt words uploaded on the intelligent development platform; wherein, The large language model includes a question-answering language model or a reasoning language model provided by the intelligent development platform.

8. The resource processing method for a live broadcast program according to claim 7, further comprising: According to the debugging data input into the intelligent development platform, the generated intelligent agent calls the large language model to perform resource interactive recommendation to obtain the debugging result; The debugging result is used to adjust the configuration information of the generated intelligent agent or the prompt word template.

9. The resource processing method for a live broadcast program according to claim 1, wherein determining the resource interaction type of the resource request based on the resource interaction record of the accessing user comprises: Detecting whether there is a resource interaction record of the live broadcast program in the resource flow record of the application program of the accessing user running the live broadcast program; If so, determine that the resource interaction type is the preset interaction type.

10. The resource processing method for a live broadcast program according to claim 1, wherein the resource interaction processing according to the target resource value submitted by the user terminal comprises: The virtual resources of the live broadcast program are purchased according to the target resource value, and the purchased virtual resources are transferred to the program account of the accessing user in the live broadcast program; wherein the payment resources for purchasing the virtual resources are deducted from the application account of the application running the live broadcast program.

11. A resource processing method for a live broadcast program, comprising: Generate resource requests for resource interaction based on the resource interaction operations submitted by the accessing user in the live broadcast program and submit them to the server; Receive and display the resource interaction value returned by the server; The resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target resource value selected by the accessing user from the resource interaction value is obtained and submitted to the server, so as to perform resource interaction processing according to the target resource value.

12. The resource processing method of a live broadcast program according to claim 11, wherein the resource interaction operation comprises a triggering operation on a resource interaction interface of the live broadcast program; The receiving and displaying the resource interaction value for resource interaction returned by the server includes: A resource interaction value for resource interaction returned by the server is received, and the resource interaction value is displayed on a resource interaction interface displayed after the resource interaction interface is triggered.

13. The resource processing method for a live broadcast program according to claim 11, wherein the agent calls a large language model to perform interactive resource recommendation, comprising: Creating an asynchronous task through the agent and submitting the asynchronous task to the thread pool for task thread allocation; The asynchronous task is executed by the agent in the assigned task thread, and the large language model is called during the execution of the asynchronous task.

14. The resource processing method for a live broadcast program according to claim 13, wherein the execution of the asynchronous task comprises: generating a prompt text according to a prompt word template and the resource interaction value recorded in the read resource interaction record; The prompt text is input into the large language model for resource interaction recommendation to obtain the resource interaction value.

15. A resource processing device for a live broadcast program, comprising: A request receiving module configured to receive a resource request submitted by a user terminal for accessing a resource interaction performed by the user in the live broadcast program; an interaction type determination module, configured to determine the resource interaction type of the resource request according to the resource interaction record of the accessing user; a value reading module configured to read, if the resource interaction type is a preset interaction type, a resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction module is configured to return the resource interaction value to the user terminal and perform resource interaction processing according to the target resource value submitted by the user terminal.

16. A resource processing device for a live broadcast program, comprising: The request submission module is configured to generate a resource request for resource interaction according to the resource interaction operation submitted by the accessing user in the live broadcast program and submit it to the server; a value display module, configured to receive and display the resource interaction value of the resource interaction returned by the server; The resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target value submission module is configured to obtain the target resource value selected by the accessing user from the resource interaction values and submit it to the server so as to perform resource interaction processing according to the target resource value.

17. A server comprising: processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to: Receive a resource request submitted by a user terminal for accessing resources for interaction with the user in the live broadcast program; Determining the resource interaction type of the resource request according to the resource interaction record of the accessing user; If the resource interaction type is a preset interaction type, read the resource interaction value obtained and stored by the agent calling the large language model to perform resource interaction recommendation; The resource interaction value is returned to the user terminal, and resource interaction processing is performed according to the target resource value submitted by the user terminal.

18. A user terminal, comprising: processor; and a memory configured to store computer-executable instructions that, when executed, cause the processor to: Generate resource requests for resource interaction based on the resource interaction operations submitted by the accessing user in the live broadcast program and submit them to the server; Receive and display the resource interaction value returned by the server; The resource interaction value is obtained by the intelligent agent calling the large language model to perform resource interaction recommendation; The target resource value selected by the accessing user from the resource interaction value is obtained and submitted to the server, so as to perform resource interaction processing according to the target resource value.

19. A computer-readable storage medium for storing computer-executable instructions, wherein the computer-executable instructions implement the steps of the method according to claim 1 or 11 when executed.

Images