Business processing method and apparatus, electronic device, and storage medium

Abstract

Embodiments of the present application disclose a service processing method and device, electronic equipment and a storage medium. The method comprises: in response to a target music capability invocation request initiated by an agent invoker, obtaining original music service data returned after execution of a target music capability; determining a target interaction semantic template corresponding to the original music service data from a plurality of interaction semantic templates based on a capability type corresponding to the original music service data; writing a target field in the original music service data into a same executable interaction object according to a mapping rule included in the target interaction semantic template to generate a target executable interaction object; and returning the target executable interaction object to the agent invoker, so that the agent invoker completes display and subsequent action execution based on the target executable interaction object without local hard-coded mapping. Through the above method, the terminal side can directly complete display and subsequent action invocation without additional analysis and derivation.

Description

Technical Field

[0001] This application belongs to the field of business processing technology, specifically relating to a business processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] In scenarios where an intelligent agent invokes music capabilities, the server typically only returns the business execution result, such as a success code, a prompt message, and the original business data. If the client wants to continue with subsequent actions such as "playing, jumping, adjusting, and adding to favorites," it often needs to hardcode the UI design and action entry points based on the business type. Summary of the Invention

[0003] In view of the above problems, this application proposes a business processing method, apparatus, electronic device, and storage medium to improve the above problems.

[0004] In a first aspect, embodiments of this application provide a business processing method applied to a server. The method includes: responding to a target music capability call request initiated by an intelligent agent caller, obtaining original music business data returned after the target music capability is executed; determining a target interactive semantic template corresponding to the original music business data from multiple interactive semantic templates based on the capability type corresponding to the original music business data, wherein different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules; writing the target field in the original music business data into the same executable interactive object according to the mapping rules included in the target interactive semantic template, generating a target executable interactive object; and sending the target executable interactive object back to the intelligent agent caller, so that the intelligent agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

[0005] Secondly, embodiments of this application provide a business processing method applied to an intelligent agent caller. The method includes: receiving a target executable interactive object returned by a server, wherein the target executable interactive object is a target music capability call request initiated by the intelligent agent caller in response to the server; obtaining the original music business data returned after the execution of the target music capability; determining a target interactive semantic template corresponding to the original music business data from multiple interactive semantic templates based on the capability type corresponding to the original music business data; and generating the target interactive object by writing the target field in the original music business data into the same executable interactive object according to the mapping rules included in the target interactive semantic template, wherein different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules; and completing the display and subsequent action execution based on the target executable interactive object without performing local hard-coding mapping.

[0006] Thirdly, embodiments of this application provide a business processing apparatus running on a server. The apparatus includes: an acquisition unit, configured to acquire, in response to a target music capability call request initiated by an intelligent agent caller, the original music business data returned after the target music capability is executed; a determination unit, configured to determine, based on the capability type corresponding to the original music business data, a target interactive semantic template corresponding to the original music business data from multiple interactive semantic templates, wherein different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules; a writing unit, configured to write the target field in the original music business data into the same executable interactive object according to the mapping rules included in the target interactive semantic template, thereby generating a target executable interactive object; and a return unit, configured to return the target executable interactive object to the intelligent agent caller, so that the intelligent agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coding mapping.

[0007] Fourthly, embodiments of this application provide an electronic device, including one or more processors and a memory; one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the one or more programs are configured to perform the methods described above.

[0008] Fifthly, embodiments of this application provide a computer-readable storage medium storing program code, wherein the above-described method is executed when the program code is run.

[0009] Sixthly, embodiments of this application provide a computer program product, including a computer program / instructions, which, when executed by a processor, implement the steps of the above-described method.

[0010] This application provides a business processing method, apparatus, electronic device, and storage medium. First, in response to a target music capability call request initiated by an intelligent agent caller, the original music business data returned after the target music capability is executed is obtained. Then, based on the capability type of the target music capability, a target interactive semantic template corresponding to the original music business data is determined from multiple interactive semantic templates. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules. According to the mapping rules included in the target interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate a target executable interactive object. Finally, the target executable interactive object is returned to the intelligent agent caller, enabling the intelligent agent caller to complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping. Through this method, after the client obtains the integrated interactive object, it can directly complete the display and subsequent action call without additional parsing and derivation, reducing the computational overhead and request chain time on the client side, significantly improving the response speed of music-related interactions, and providing a smoother user experience. Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A flowchart of a business processing method according to an embodiment of this application is shown; Figure 2 A flowchart of a business processing method according to another embodiment of this application is shown; Figure 3 A schematic diagram of the server structure in another embodiment of this application is shown; Figure 4 A flowchart of a business processing method according to another embodiment of this application is shown; Figure 5 This paper shows a structural block diagram of a service processing apparatus according to an embodiment of the present application; Figure 6 This paper shows a structural block diagram of a service processing apparatus according to an embodiment of the present application; Figure 7 This paper shows a structural block diagram of an electronic device for performing a service processing method according to an embodiment of this application. Figure 8This application illustrates a storage unit for storing or carrying program code that implements the business processing method according to the embodiments of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0014] In scenarios where an intelligent agent invokes music capabilities, the server typically only returns the business execution result, such as a success code, a prompt message, and the original business data. If the client wants to continue with subsequent actions such as "playing, jumping, adjusting, and adding to favorites," it often needs to hardcode the UI design and action entry points based on the business type.

[0015] This approach has three problems: The client must pre-configure the follow-up action logic for each business result, resulting in high coupling. The separation of the server-side response and the entry point for subsequent client actions makes it difficult to guarantee consistency. When capabilities are expanded or page navigation is adjusted, both the server and client need to be modified synchronously.

[0016] Therefore, simply returning business data cannot meet the requirement of the edge-side intelligent agent to "continue to execute the next action after seeing the result".

[0017] Therefore, the inventors have proposed the business processing method, apparatus, server, and storage medium in this application. First, in response to a target music capability call request initiated by the intelligent agent caller, the original music business data returned after the target music capability is executed is obtained. Then, based on the capability type of the target music capability, the target interactive semantic template corresponding to the original music business data is determined from multiple interactive semantic templates. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules. According to the mapping rules included in the target interactive semantic template, the target fields in the original music business data are written into the same executable interactive object, generating a target executable interactive object. Finally, the target executable interactive object is returned to the intelligent agent caller, enabling the intelligent agent caller to complete the display and subsequent action execution based on the target executable interactive object without local hard-coding mapping. Through this method, after the client-side obtains the integrated interactive object, it can directly complete the display and subsequent action calls without additional parsing and derivation, reducing the interaction frequency between the client and server, reducing the computational overhead and request chain time on the client-side, significantly improving the response speed of music-related interactions, and providing a smoother user experience.

[0018] The embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0019] Please see Figure 1 This application provides a business processing method applied to a server, the method comprising: Step S110: In response to the target music capability call request initiated by the intelligent agent caller, obtain the original music service data returned after the target music capability is executed.

[0020] In this embodiment, the intelligent agent caller, located on the calling side, is an LLM / Agent program entity that acts as a relay scheduler between the end-side intelligent agent and the semantic mapping server. It receives music-related natural language commands from the end-side user, generates standardized tool call requests after completing intent understanding, and sends them to the capability access layer of the semantic mapping server. At the same time, it also receives the next round of action call requests triggered by subsequent executable interactive objects, and completes secondary scheduling.

[0021] The target music capability call request is a standardized tool call instruction generated by the intelligent agent caller based on the user's music intent, and it is a trigger signal that triggers the execution of the underlying music service.

[0022] Specifically, after the intelligent agent caller receives a user's music-related instructions (such as searching for songs, querying playlists, adjusting playback progress, or adding music to favorites), the associated LLM / Agent caller generates a corresponding music capability tool call request to complete the initial triggering action. After the LLM / Agent caller sends the generated tool call request to the capability access layer of the semantic mapping server, the capability access layer will send an execution instruction to the corresponding underlying music business module. After the underlying music business module completes the corresponding business logic processing, it can return the corresponding raw music business data.

[0023] The raw music business data is returned by the underlying business module that provides core music capabilities. It is the original return content output by the underlying music business module after completing the corresponding capability logic execution. Specifically, it includes three types of content: the status indicator of business execution (such as execution success / failure code, corresponding business type mark), the original prompt text for display, and the raw business data that has not undergone structured processing.

[0024] In this application embodiment, common music tool call requests that can trigger this process cover various music business scenarios preset in the solution, including: search_music song search tool calls, get_play_list / get_music_resource playlist / music resource acquisition tool calls, query_playback_state / control_volume playback status query / volume control tool calls, favorite_song song collection tool calls, etc. Any of the above-mentioned legitimate music capability tool call requests can completely trigger the subsequent return process of raw music business data, entering the semantic mapping stage.

[0025] In this embodiment, after receiving a user's music-related instruction, the edge agent transmits the natural language intent to the associated LLM / Agent program. The LLM / Agent program parses the intent, matches the corresponding callable music tool capabilities, generates a compliant target music capability call request, and completes the standardized translation of the instruction. Then, the agent sends the generated target music capability call request to the capability access layer of the semantic mapping server. After verifying the request's validity, the capability access layer routes and distributes it to the corresponding underlying music service execution module based on the capability identifier carried in the request. Upon receiving the execution instruction, the underlying music service module completes the corresponding logic processing, such as performing song searches, retrieving a specified user's playlist, adjusting the current playback volume, and updating song collection status. It then packages and returns the completed status code, prompt text, and original business data, which is received by the capability access layer of the semantic mapping server. This completes the entire process of "obtaining original music business data," and the module will subsequently enter the downstream mapping stage of capability recognition and semantic template matching.

[0026] Optionally, in this embodiment, if the server receives multiple raw music service data, it will run a complete result-action mapping process for each independent raw music service data separately: identify the capability type corresponding to each raw music service data, match the corresponding interactive semantic template, extract the respective state / display / action parameter fields, generate an independent executable interactive object for each raw music service data, and finally package all generated interactive objects and send them back to the intelligent agent caller in a unified manner, so that there will be no problem of multiple service results being confused or fields being disordered.

[0027] Step S120: Based on the capability type corresponding to the original music service data, determine the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates, wherein different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules.

[0028] In this embodiment, the capability type is a standardized classification identifier for all music business capabilities. It is a classification of different music capabilities based on the core interactive attributes, execution action types, and return result forms of music business. For example, jump jump type, resource acquisition type, progress / quantity adjustment type, state switching type, etc. Each type of capability has a unique identifier, which is used to quickly match the corresponding interactive semantic template in the future.

[0029] Interactive semantic templates are standardized dual semantic constraint frameworks predefined on the server side. Different templates correspond to different music interaction scenarios. Each type of template has a predefined exclusive field structure and carries the definition specifications of display rendering rules and subsequent action entry points. It is a unified rule carrier for mapping raw business data to executable interactive objects, such as jump jump templates, parameter-list resource list templates, process progress adjustment templates, and switch status switch templates.

[0030] The target interaction semantic template refers to a unique template that is selected and matched from all predefined interaction semantic templates for a specific original music business data. All subsequent field extraction and interaction object assembly actions will strictly follow the rules of this template.

[0031] The mapping rules are the exclusive field transformation logic defined within each interactive semantic template. They clearly stipulate how to extract status fields, display fields, and action entry parameters from the original music business data, and how these fields are assembled into an integrated executable interactive object that contains both display and execution semantics according to the template's fixed structure. The field extraction logic and action entry definition methods of different templates are completely different.

[0032] In this embodiment, the server calls the capability type identification module, which uses the unique identifier of the music capability interface being called and the service type mark carried in the original music service data to double-verify and confirm the current capability type, thus avoiding capability type identification errors.

[0033] Specifically, the server first reads the list of default candidate templates corresponding to the capability type from the configuration center, and then selects the set of candidate templates that can be adapted to the capability type from all four predefined interactive semantic templates to complete the first round of range convergence.

[0034] Then, the server parses the returned resource format and execution status fields of the original music business data, verifies each template in the candidate set to see if it matches the interaction attributes of the current business result, and eliminates all obviously mismatched candidate templates.

[0035] If multiple candidate templates remain after verification, the server can select the template that best fits the current business scenario according to the preset priority logic of "result interaction form matching degree takes precedence over capability default mapping rule", and finally determine the unique target interaction semantic template.

[0036] Step S130: According to the mapping rules included in the target interaction semantic template, write the target fields in the original music business data into the same executable interaction object to generate the target executable interaction object.

[0037] In this embodiment, the target fields are three types of core fields extracted from the original music business data: result status fields that mark the business execution status, display semantic fields used for terminal rendering and display, and action entry fields that carry the rules for subsequent action calls. These are the core materials for assembling executable interactive objects.

[0038] An executable interactive object is an integrated structured object that carries both display and execution semantics. Once the client obtains this object, it can directly complete the result display without needing to hardcode additional rules locally. At the same time, it can read the action parameters within the object to initiate the next round of music business calls, achieving a seamless connection between "result display" and "subsequent actions".

[0039] Specifically, the server-side field extraction module loads the mapping rules of the target interaction semantic template, accurately extracts all target fields from the original music business data according to the field mapping relationship defined by the rules, writes the three types of target fields into the corresponding field positions of the same interaction object according to the fixed structured format preset by the template, and completes the structured assembly of the fields, performs a legality check on the assembled object, confirms that all required fields are complete and the format conforms to the specifications, and finally generates a target executable interaction object that meets the requirements and sends it back to the intelligent agent caller.

[0040] Step S140: The target executable interactive object is sent back to the intelligent agent caller so that the intelligent agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

[0041] In this embodiment, the server-side return module first performs a validity check on the generated target executable interactive object to confirm that all required fields are complete and the format conforms to the general protocol specifications of the client side. Then, it encapsulates the object into a structured message that conforms to the intelligent agent caller's transmission protocol.

[0042] The server sends the encapsulated message back to the intelligent agent caller, and at the same time writes all the parameters of the target executable object sent back this time into the audit log for subsequent problem backtracking and mapping rule iteration.

[0043] The intelligent agent caller passes the target executable interactive object to the general rendering component on the client side. The component directly reads the UI type identifier in the object, automatically matches the corresponding preset general UI template, and completes the display of the music business results. No business-specific display rules need to be hard-coded locally throughout the process.

[0044] The client-side action execution layer directly reads the action entry field within the target executable object, automatically assembles the request message for the next round of music service calls, and initiates a new tool call request to the server. It can complete subsequent actions such as jumping, adjusting, and adding to favorites without the need for local hard-coding of action mapping logic.

[0045] This application provides a business processing method that completes the entire process from business result generation to semantic mapping on the server side, completely avoiding the misalignment problem between display logic and execution logic caused by the traditional model of "server returning raw data and client side deriving action rules". This significantly improves the accuracy of interaction. After the client side obtains the integrated interaction object, it can directly complete the display and subsequent action calls without additional parsing and derivation, reducing the computational overhead and request chain time on the client side, significantly improving the response speed of music-related interactions and making the user experience smoother.

[0046] Please see Figure 2 This application provides a business processing method applied to a server, the method comprising: Step S210: In response to the target music capability call request initiated by the intelligent agent caller, obtain the original music service data returned after the target music capability is executed.

[0047] Step S220: Based on the capability type corresponding to the original music service data, determine the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates, wherein different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules.

[0048] In this embodiment, the plurality of interactive semantic templates include a first interactive semantic template, a second interactive semantic template, a third interactive semantic template, and a fourth interactive semantic template. The first interactive semantic template indicates that the action entry point is a jump address protocol, and the intelligent agent caller does not need to call other interfaces; it can directly read the jump data to complete the page jump. The second interactive semantic template indicates that the action entry point is a specified interface bound to each list item. After the intelligent agent caller selects any list item, it can read the parameters in the corresponding list item data structure and call the list request interface to complete the corresponding action. The third interactive semantic template indicates that the action entry point is a progress / quantity adjustment interface. The intelligent agent caller can read the current progress value to display the current progress. When dragging the progress bar, it automatically calls the progress query / control interface to pass in the adjusted target value to complete the quantity / progress update. The fourth interactive semantic template indicates that the action entry point is the corresponding interface of two binary states. The intelligent agent caller reads the current switch state and renders the switch state. When clicking the switch to switch states, it calls the corresponding open operation interface or close operation interface to complete the state change.

[0049] In this embodiment of the application, the first interactive semantic template refers to the jump jump class template, the second interactive semantic template refers to the parameter-list resource list class template, the third interactive semantic template refers to the process progress adjustment class template, and the fourth interactive semantic template refers to the switch state switch class template.

[0050] It is clear that the differences in rules among different interactive semantic templates are not simply differences in display style, but rather execution semantic constraints for different interactive scenarios. The core differences lie in three dimensions: field composition, the definition logic of action entry points, and the execution logic on the client side. Jump redirect template: Used for redirect results. Core fields include ui_card_type and jump_data. The rule is defined as the action entry point being the redirect address protocol. The client does not need to call any other interfaces; it can directly read jump_data to complete the page redirection.

[0051] Parameter-list resource list template: Used for candidate lists or resource lists. Core fields include ui_card_type, parameter_list_api, and list_item_json. The rule definition is that the action entry point is a specified interface bound to each list item. After selecting any list item on the client side, the parameters in the corresponding list_item_json can be read to call parameter_list_api to complete the corresponding action.

[0052] Process progress adjustment class template: Used for progress or quantity adjustment. The core fields include ui_card_type, process_bar_api, and process_bar_value. The rule definition is that the action entry is the progress / quantity adjustment interface. The client can read the current process_bar_value to display the current progress. When the progress bar is dragged, the process_bar_api is automatically called to pass in the adjusted target value to complete the quantity / progress update.

[0053] Switch status switch template: Used for binary status switching. Core fields include ui_card_type, switch_status, switch_open_api, and switch_close_api. The rule is defined as the action entry point being the corresponding interface of the two binary states. The client reads the current switch_status to render the switch status. When the switch is clicked to switch the status, the corresponding switch_open_api or switch_close_api is called to complete the status change.

[0054] As one approach, determining the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates based on the capability type corresponding to the original music service data includes: if the capability type of the target music capability is a jump type, determining the target interactive semantic template corresponding to the original music service data as the first interactive semantic template; if the capability type of the target music capability is a resource acquisition type, determining the target interactive semantic template corresponding to the original music service data as the second interactive semantic template; if the capability type of the target music capability is a progress or quantity adjustment type, determining the target interactive semantic template corresponding to the original music service data as the third interactive semantic template; and if the capability type of the target music capability is a state switching type, determining the target interactive semantic template corresponding to the original music service data as the fourth interactive semantic template.

[0055] In this embodiment of the application, the initialization and entry of four sets of strong binding relationships can be completed in advance in the configuration center of the semantic mapping server: Jump-type capabilities (such as search_music, identify_music, open_music_page) correspond to the first interactive semantic template; Resource acquisition capabilities (such as get_play_list, get_music_resource) correspond to the second interactive semantic template; Progress / quantity adjustment capabilities (such as query_playback_state, control_volume) correspond to the third interaction semantic template; State-switching capabilities (such as favorite_song) correspond to the fourth interaction semantic template.

[0056] At the same time, configure the corresponding field extraction rules and mapping verification logic for each binding relationship to ensure that the matching process is unambiguous.

[0057] The server uses a capability type identification module to double-check and confirm the category of the current capability by combining the unique identifier of the interface being called and the business type mark in the original music business data, thus avoiding capability type identification errors.

[0058] Specifically, the server reads the predefined strong binding mapping list in the configuration center, and directly matches the corresponding unique interaction semantic template based on the confirmed capability type. This eliminates the need for complex logic such as multiple candidate template filtering and secondary verification, allowing the server to directly obtain the target interaction semantic template.

[0059] Step S230: Extract the target result status field, target display field, and target action entry parameter from the original music service data. The target result status field is used to indicate the execution status and type of the target music capability. The target display field is used to describe the display content and / or display rules of the original music service data. The target action entry parameter is used to indicate the rules for carrying subsequent action calls.

[0060] In this embodiment, the server-side field extraction module performs structured parsing of the original music business data according to the mapping rules of the determined target interaction semantic template: first, it extracts the status code and business type marker returned by the business as the target result status field; then, it extracts the prompt text, resource cover, name, and other content used for front-end display as the target display semantic field; finally, it parses the jump protocol, list operation interface, progress / quantity adjustment interface, and other action parameters from the original data as the target action entry field. The three types of fields together form the target field set.

[0061] Step S240: According to the mapping rules included in the target interaction semantic template, write the target result status field, target display field and target action entry parameter into the same executable interaction object to generate a target executable interaction object.

[0062] As one approach, the step of writing target fields from the original music service data into the same executable interaction object according to the mapping rules included in the target interaction semantic template to generate a target executable interaction object includes: writing target fields from the original music service data into the same executable interaction object according to the mapping rules included in the first interaction semantic template to generate a first executable interaction object; or writing target fields from the original music service data into the same executable interaction object according to the mapping rules included in the second interaction semantic template to generate a second executable interaction object; or writing target fields from the original music service data into the same executable interaction object according to the mapping rules included in the third interaction semantic template to generate a third executable interaction object; or writing target fields from the original music service data into the same executable interaction object according to the mapping rules included in the fourth interaction semantic template to generate a fourth executable interaction object.

[0063] In this embodiment of the application, after the target interaction semantic template is determined in the manner described above, the target result status field, target display field and target action entry parameter extracted from the original music business data can be written into the same executable interaction object according to the mapping rules corresponding to the target interaction semantic template, so as to generate an executable interaction object in the format of the target interaction semantic template.

[0064] Specifically, three target fields—search status code, search result prompt text, and redirection protocol—are extracted from the raw data of song search. These fields are then assembled according to the jump redirection template rules to generate the first executable interactive object. The client side can directly read the redirection protocol to open the search results page.

[0065] Extract the playlist retrieval status code, playlist cover / name display field, and playback interface parameters bound to list items from the raw data of the playlist resource retrieval class. Assemble them according to the parameter-list resource list class template rules to generate a second executable interactive object. The client can directly render the playlist list, and clicking any item will automatically call the playback interface.

[0066] Extract the adjustment status code, current progress value display field, and progress / quantity adjustment interface parameters from the original data of the playback progress / quantity adjustment class. Assemble them according to the process progress adjustment class template rules to generate a third executable interactive object. The progress bar can be directly rendered on the client side, and the adjustment interface is automatically called to update the playback progress when dragged.

[0067] Extract the switching status code, current collection status display field, and status switching interface parameters from the original data of the collection status switching class. Assemble them according to the switch status switch class template rules to generate a fourth executable interactive object. The client can directly render the collection switch, and when clicked, the switching interface will be automatically called to update the collection status.

[0068] Step S250: The target executable interactive object is sent back to the intelligent agent caller so that the intelligent agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

[0069] For example, such as Figure 3 As shown, the server in this embodiment of the application includes at least: Capability type identification module: Identifies the capability type corresponding to the current business result; Semantic template selection module: Selects the target template from multiple interactive semantic templates; Result shaping module: Extracts and displays semantic fields and action entry fields; Interaction object generation module: Generates a unified executable interaction object; Postback module: Returns the interaction object to the caller.

[0070] The process described in steps S210-S250 of this embodiment can be as follows: Taking a user's request to "search for a specific song" as an example, the complete process is as follows: The LLM / Agent call direction semantic mapping server initiates a search_music search tool call request.

[0071] The server-side capability access layer receives the raw business data indicating a successful search from the music search capability, enters the capability type identification module, and determines that the current capability type is search_music.

[0072] The semantic template selection module matches the corresponding jump redirection template, and the field extraction module extracts the search success status marker, search result display text, and action parameters such as the redirection target protocol address allmusic: / / search?xxx from the original search results.

[0073] The interaction object generation module assembles and generates an executable interaction object according to the jump jump class template rules, which includes apiproxy_biz_ui_card_type = JUMP_PAGE, apiproxy_biz_jump_data = allmusic: / / search?..., and also carries a result status field for successful search.

[0074] The integrated object is sent back to the client. After the client's presentation layer recognizes that ui_card_type is JUMP_PAGE, it renders the search result suggestion card. The action execution layer directly reads the protocol in jump_data and can directly open the corresponding search result page without the need for local preset redirection rules, thus completing the subsequent redirection action.

[0075] This application provides a business processing method that can unify various diverse music business capabilities under multiple preset template frameworks, allowing different types of music businesses to correspond to standardized structured rules, avoiding the problem of scattered business logic being unable to be managed in a unified manner, and supporting the server to quickly complete the semantic mapping of business results.

[0076] Please see Figure 4 This application provides a service processing method applied to an electronic device, the method comprising: Step S310: Receive the target executable interaction object returned by the server. The target executable interaction object is the original music service data returned by the server after the execution of the target music service data in response to the target music service data call request initiated by the intelligent agent. Based on the capability type corresponding to the original music service data, the server determines the target interaction semantic template corresponding to the original music service data from multiple interaction semantic templates. Then, according to the mapping rules included in the target interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate the object. Different capability types correspond to different interaction semantic templates, and different interaction semantic templates include different mapping rules.

[0077] In this embodiment of the application, the electronic device is a client that includes an intelligent agent caller.

[0078] Step S320: Based on the target executable interactive object, complete the display and subsequent action execution without performing local hard-coded mapping.

[0079] As one approach, the process of completing the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping includes: identifying the current interaction mode based on the identifier card type within the target executable interactive object, calling the corresponding general rendering component to complete the result display; and completing the subsequent action execution based on the interface class action parameters within the target executable interactive object.

[0080] In this embodiment, after the intelligent agent caller receives the executable interactive object returned by the server, it does not need to hardcode any business action rules locally. It completes the subsequent action execution in two steps: First, it identifies the current interaction form through the ui_card_type in the target executable interactive object and calls the corresponding general rendering component to complete the result display; Second, according to the action entry field convention of different templates, it directly reads the *_api class action parameters in the target executable interactive object, automatically assembles the request message for the next round of tool call, and initiates a new tool call request to the LLM / Agent caller to complete subsequent actions such as playback, jump, adjustment, and favorite.

[0081] Taking the "song search" service as an example: After obtaining the executable interactive object of the jump class on the client side, it directly recognizes the UI identifier of JUMP_PAGE, automatically renders a search success prompt card, and reads the jump protocol within the object to directly open the corresponding search results page.

[0082] This application provides a business processing method in which the business mapping logic converges on the server side. The client side does not need to pre-define exclusive display and action mapping rules for each type of music business. Business consumption can be completed using only general components, which greatly reduces the maintenance cost and iteration threshold of the client side code.

[0083] Please see Figure 5 This application provides a business processing device 400, which operates on a server. The device 400 includes: The acquisition unit 410 is used to respond to a target music capability call request initiated by the intelligent agent caller and acquire the original music service data returned after the target music capability is executed.

[0084] The determining unit 420 is used to determine the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates based on the capability type corresponding to the original music service data. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules.

[0085] In this embodiment, the plurality of interactive semantic templates include a first interactive semantic template, a second interactive semantic template, a third interactive semantic template, and a fourth interactive semantic template. The first interactive semantic template indicates that the action entry point is a jump address protocol, and the intelligent agent caller does not need to call other interfaces; it can directly read the jump data to complete the page jump. The second interactive semantic template indicates that the action entry point is a specified interface bound to each list item. After the intelligent agent caller selects any list item, it can read the parameters in the corresponding list item data structure and call the list request interface to complete the corresponding action. The third interactive semantic template indicates that the action entry point is a progress / quantity adjustment interface. The intelligent agent caller can read the current progress value to display the current progress. When dragging the progress bar, it automatically calls the progress query / control interface to pass in the adjusted target value to complete the quantity / progress update. The fourth interactive semantic template indicates that the action entry point is the corresponding interface of two binary states. The intelligent agent caller reads the current switch state and renders the switch state. When clicking the switch to switch states, it calls the corresponding open operation interface or close operation interface to complete the state change.

[0086] In one approach, the determining unit 420 is specifically configured to: if the capability type of the target music capability is a jump type, determine the target interaction semantic template corresponding to the original music service data as the first interaction semantic template; if the capability type of the target music capability is a resource acquisition type, determine the target interaction semantic template corresponding to the original music service data as the second interaction semantic template; if the capability type of the target music capability is a progress or quantity adjustment type, determine the target interaction semantic template corresponding to the original music service data as the third interaction semantic template; and if the capability type of the target music capability is a state switching type, determine the target interaction semantic template corresponding to the original music service data as the fourth interaction semantic template.

[0087] The writing unit 430 is used to write the target fields in the original music business data into the same executable interaction object according to the mapping rules included in the target interaction semantic template, thereby generating the target executable interaction object.

[0088] In one approach, the writing unit 430 is specifically used to extract a target result status field, a target display field, and a target action entry parameter from the original music service data. The target result status field is used to indicate the execution status and type of the target music capability, the target display field is used to describe the display content and / or display rules of the original music service data, and the target action entry parameter is used to indicate the rules for carrying subsequent action calls. According to the mapping rules included in the target interaction semantic template, the target result status field, the target display field, and the target action entry parameter are written into the same executable interaction object to generate a target executable interaction object.

[0089] Alternatively, according to the mapping rules included in the first interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate a first executable interaction object; or, according to the mapping rules included in the second interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate a second executable interaction object; or, according to the mapping rules included in the third interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate a third executable interaction object; or, according to the mapping rules included in the fourth interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate a fourth executable interaction object.

[0090] The return unit 440 is used to return the target executable interactive object to the intelligent agent caller, so that the intelligent agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

[0091] Please see Figure 6 This application provides a service processing device 500, which operates in an electronic device. The device 500 includes: The receiving unit 510 is used to receive a target executable interactive object returned by the server. The target executable interactive object is generated by the server responding to the target music capability call request initiated by the intelligent agent caller, obtaining the original music service data returned after the target music capability is executed, determining the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates based on the capability type corresponding to the original music service data, and generating the target interactive object by writing the target fields in the original music service data into the same executable interactive object according to the mapping rules included in the target interactive semantic template. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules.

[0092] The execution unit 520 is used to complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

[0093] In one approach, the execution unit 520 is specifically used to identify the current interaction mode based on the identifier card type within the target executable interactive object, call the corresponding general rendering component to complete the result display, and complete the subsequent action execution based on the interface class action parameters within the target executable interactive object.

[0094] It should be noted that the device embodiments in this application correspond to the aforementioned method embodiments. The specific principles in the device embodiments can be found in the content of the aforementioned method embodiments, and will not be repeated here.

[0095] The following will combine Figure 7 This application describes an electronic device.

[0096] Please see Figure 7 Based on the aforementioned business processing methods and apparatus, this application embodiment also provides another electronic device 800 capable of executing the aforementioned business processing methods. The electronic device 800 includes one or more (only one shown in the figure) processors 802 and a memory 804 coupled together. The memory 804 stores programs capable of executing the contents of the aforementioned embodiments, and the processors 802 can execute the programs stored in the memory 804.

[0097] The processor 802 may include one or more processing cores. The processor 802 connects to various parts within the electronic device 800 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 804, and by calling data stored in the memory 804. Optionally, the processor 802 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 802 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the displayed content; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 802 and may be implemented separately using a communication chip.

[0098] The memory 804 may include random access memory (RAM) or read-only memory (ROM). The memory 804 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 804 may include a program storage area and a data storage area. The program storage area may store instructions for implementing server operations, instructions for implementing at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), and instructions for implementing the various method embodiments described below. The data storage area may also store data created by the electronic device 800 during use (such as phonebook data, audio and video data, chat log data, etc.).

[0099] Please refer to Figure 8 This diagram illustrates a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. The computer-readable storage medium 900 stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0100] The computer-readable storage medium 900 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 900 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 900 has storage space for program code 910 that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code 910 may, for example, be compressed in a suitable form.

[0101] This application provides a business processing method, apparatus, electronic device, and storage medium. First, in response to a target music capability call request initiated by an intelligent agent caller, the method obtains the original music business data returned after the target music capability is executed. Then, based on the capability type of the target music capability, it determines the target interactive semantic template corresponding to the original music business data from multiple interactive semantic templates. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules. According to the mapping rules included in the target interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate a target executable interactive object. Finally, the target executable interactive object is returned to the intelligent agent caller, enabling the intelligent agent caller to complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping. Through this method, after the client obtains the integrated interactive object, it can directly complete the display and subsequent action call without additional parsing and derivation, reducing the computational overhead and request chain time on the client side, significantly improving the response speed of music-related interactions, and providing a smoother user experience.

[0102] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A business processing method, characterized in that, Applied to the server side, the method includes: In response to a target music capability call request initiated by an intelligent agent caller, the original music service data returned after the target music capability is executed is obtained; Based on the capability type corresponding to the original music service data, a target interactive semantic template corresponding to the original music service data is determined from multiple interactive semantic templates. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules. The multiple interactive semantic templates include a first interactive semantic template, a second interactive semantic template, a third interactive semantic template, and a fourth interactive semantic template. The first interactive semantic template is used to indicate that the action entry point is a jump address protocol. The intelligent agent caller does not need to call other interfaces. It can directly read the jump data to complete the page jump. The second interactive semantic template is used to indicate that the action entry point is a specified interface bound to each list item. After the intelligent agent selects any list item, it can read the parameters in the corresponding list item data structure and call the list request interface to complete the corresponding action. The third interactive semantic template is used to indicate that the action entry is the progress / quantity adjustment interface. The intelligent agent can read the current progress value and display the current progress. When dragging the progress bar, the progress query / control interface is automatically called to pass in the adjusted target value to complete the quantity / progress update. The fourth interactive semantic template is used to indicate that the action entry is the corresponding interface of two binary states. The intelligent agent caller reads the current switch state and renders the switch state. When the switch is clicked to switch the state, the corresponding open operation interface or close operation interface is called to complete the state change. According to the mapping rules included in the target interaction semantic template, the target fields in the original music business data are written into the same executable interaction object to generate the target executable interaction object; The target executable interactive object is sent back to the agent caller, so that the agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

2. The method according to claim 1, characterized in that, The step of writing the target fields from the original music business data into the same executable interaction object according to the mapping rules included in the target interaction semantic template, and generating the target executable interaction object, includes: Extract the target result status field, target display field, and target action entry parameter from the original music service data. The target result status field is used to indicate the execution status and type of the target music capability. The target display field is used to describe the display content and / or display rules of the original music service data. The target action entry parameter is used to indicate the rules for carrying subsequent action calls. According to the mapping rules included in the target interaction semantic template, the target result status field, target display field, and target action entry parameter are written into the same executable interaction object to generate a target executable interaction object.

3. The method according to claim 1, characterized in that, The step of determining the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates based on the capability type corresponding to the original music service data includes: If the capability type of the target music capability is a jump type, the target interaction semantic template corresponding to the original music service data is determined to be the first interaction semantic template; If the capability type of the target music capability is resource acquisition, the target interaction semantic template corresponding to the original music business data is determined to be the second interaction semantic template. If the capability type of the target music capability is progress or quantity adjustment, the target interactive semantic template corresponding to the original music business data is determined to be the third interactive semantic template. If the capability type of the target music capability is a state switching type, the target interaction semantic template corresponding to the original music service data is determined to be the fourth interaction semantic template.

4. The method according to claim 3, characterized in that, The step of writing the target fields from the original music business data into the same executable interaction object according to the mapping rules included in the target interaction semantic template, and generating the target executable interaction object, includes: According to the mapping rules included in the first interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate the first executable interactive object; or, According to the mapping rules included in the second interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate a second executable interactive object; or, According to the mapping rules included in the third interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate a third executable interactive object; or, According to the mapping rules included in the fourth interactive semantic template, the target fields in the original music business data are written into the same executable interactive object to generate the fourth executable interactive object.

5. A business processing method, characterized in that, Applied to the intelligent agent caller, the method includes: The server receives a target executable interaction object returned by the client. This target executable interaction object is the result of the server responding to a target music capability call request initiated by the agent. After obtaining the original music service data returned after the target music capability is executed, the server determines the target interaction semantic template corresponding to the original music service data from multiple interaction semantic templates based on the capability type corresponding to the original music service data. Then, according to the mapping rules included in the target interaction semantic template, the target fields in the original music service data are written into the same executable interaction object to generate the object. Different capability types correspond to different interaction semantic templates, and different interaction semantic templates include different mapping rules. The multiple interaction semantic templates include a first interaction semantic template, a second interaction semantic template, a third interaction semantic template, and a fourth interaction semantic template. The first interaction semantic template is used to indicate... The action entry point is a jump address protocol. The agent caller does not need to call other interfaces; it can directly read the jump data to complete the page jump. The second interaction semantic template indicates that the action entry point is a specified interface bound to each list item. After the agent caller selects any list item, it can read the parameters in the corresponding list item data structure and call the list request interface to complete the corresponding action. The third interaction semantic template indicates that the action entry point is a progress / quantity adjustment interface. The agent caller can read the current progress value to display the current progress. When dragging the progress bar, it automatically calls the progress query / control interface to pass in the adjusted target value to complete the quantity / progress update. The fourth interaction semantic template indicates that the action entry point is the corresponding interface of two binary states. The agent caller reads the current switch state and renders the switch state. When clicking the switch to switch the state, it calls the corresponding open operation interface or close operation interface to complete the state change. Based on the target executable interactive object, the display and subsequent action execution are completed without local hard-coding mapping.

6. The method according to claim 5, characterized in that, The process of completing the display and subsequent action execution based on the target executable interactive object without performing local hard-coding mapping includes: Based on the identification card type within the target executable interactive object, the current interaction mode is identified, and the corresponding general rendering component is called to complete the result display. The subsequent actions are executed based on the interface class action parameters within the target executable interactive object.

7. A business processing device, characterized in that, Running on the server side, the device includes: The acquisition unit is used to respond to a target music capability call request initiated by the intelligent agent caller and acquire the original music service data returned after the target music capability is executed. The determining unit is configured to determine the target interactive semantic template corresponding to the original music service data from multiple interactive semantic templates based on the capability type corresponding to the original music service data. Different capability types correspond to different interactive semantic templates, and different interactive semantic templates include different mapping rules. The multiple interactive semantic templates include a first interactive semantic template, a second interactive semantic template, a third interactive semantic template, and a fourth interactive semantic template. The first interactive semantic template indicates that the action entry point is a jump address protocol, allowing the intelligent agent caller to complete the page jump directly by reading the jump data without needing to call other interfaces. The second interactive semantic template indicates that the action entry point is a binding protocol. At the designated interface for each list item, after the agent caller selects any list item, it can read the parameters in the corresponding list item data structure and call the list request interface to complete the corresponding action; the third interaction semantic template is used to indicate that the action entry is the progress / quantity adjustment interface, the agent caller can read the current progress value to display the current progress, and when dragging the progress bar, it automatically calls the progress query / control interface to pass in the adjusted target value to complete the quantity / progress update; the fourth interaction semantic template is used to indicate that the action entry is the corresponding interface of two binary states, the agent caller reads the current switch state to render the switch state, and when clicking the switch to switch the state, it calls the corresponding open operation interface or close operation interface to complete the state change; The writing unit is used to write the target fields in the original music business data into the same executable interaction object according to the mapping rules included in the target interaction semantic template, thereby generating the target executable interaction object; The feedback unit is used to send the target executable interactive object back to the agent caller, so that the agent caller can complete the display and subsequent action execution based on the target executable interactive object without performing local hard-coded mapping.

8. An electronic device, characterized in that, It includes one or more processors and memory; one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, the one or more programs being configured to perform the method of any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code, wherein the program code, when executed by a processor, performs the method according to any one of claims 1-6.

Images