Feedback information generation method and device

Abstract

The invention provides a feedback information generation method and device, and relates to the technical field of artificial intelligence, in particular to the technical fields of large models, automatic driving, intelligent traffic and the like. According to the specific implementation scheme, the method comprises the steps of determining an application field of current query information and a target component set associated with the application field; inputting the current query information and the target component set into a large language model, and generating a function component calling flow chart; and the execution function component calls the action instruction sequence corresponding to the flow chart to generate feedback information. According to the mode, the accuracy and efficiency of generating the feedback information are effectively improved.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, specifically to technical fields such as large models, autonomous driving, and intelligent transportation, and more particularly to a method and device for generating feedback information. Background Art

[0002] After the emergence of large models, some dialogue systems attempted to incorporate their capabilities. However, because the understanding and usage paradigms of large models were not yet good enough, they needed to rely on samples for a large amount of supervised fine-tuning (SFT). Summary of the Invention

[0003] Embodiments of the present disclosure provide a method, apparatus, device, and storage medium for generating feedback information.

[0004] In a first aspect, an embodiment of the present disclosure provides a method for generating feedback information, the method comprising: determining the application field of the current query information and the target component set associated with the application field; inputting the current query information and the target component set into a large language model to generate a functional component call flowchart; executing the action instruction sequence corresponding to the functional component call flowchart to generate feedback information.

[0005] In the second aspect, an embodiment of the present disclosure provides a feedback information generating device, which includes: a determination module, an input module and an execution module, wherein the determination module is configured to determine the application field of the current query information and the target component set associated with the application field; the input module is configured to input the current query information and the target component set into a large language model to generate a functional component call flowchart; the execution module is configured to execute the action instruction sequence corresponding to the functional component call flowchart to generate feedback information.

[0006] In a third aspect, an embodiment of the present disclosure provides an electronic device comprising one or more processors; a storage device storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement a feedback information generation method as in any embodiment of the first aspect.

[0007] In a fourth aspect, an embodiment of the present disclosure provides a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the feedback information generating method of any embodiment of the first aspect.

[0008] In a fifth aspect, an embodiment of the present disclosure provides a computer program product, including a computer program, which, when executed by a processor, implements the feedback information generating method of any embodiment of the first aspect.

[0009] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 is an exemplary system architecture diagram in which the present disclosure may be applied; Figure 2 is a flow chart of an embodiment of a method for generating feedback information according to the present disclosure; Figure 3 is a flowchart of another embodiment of the feedback information generating method according to the present disclosure; Figure 4 is a schematic diagram of an application scenario of the feedback information generating method according to the present disclosure; Figure 5 is a schematic diagram of an embodiment of a feedback information generating device according to the present disclosure; Figure 6 It is a structural diagram of a computer system suitable for implementing the electronic device of the embodiment of the present disclosure. DETAILED DESCRIPTION

[0011] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0012] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0013] Figure 1 An exemplary system architecture 100 is shown to which an embodiment of the feedback information generating method of the present disclosure can be applied.

[0014] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is a medium for providing communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links or fiber optic cables.

[0015] Users can use terminal devices 101 , 102 , 103 to interact with server 105 via network 104 to receive or send messages, etc.

[0016] Terminal devices 101, 102, and 103 may be hardware or software. When terminal devices 101, 102, and 103 are software, they may be installed in the electronic devices listed above. They may be implemented as multiple software programs or software modules, or as a single software program or software module. This is not specifically limited here.

[0017] Server 105 can be a server that provides various services, for example, determining the application field of the current query information and the target component set associated with the application field; inputting the current query information and the target component set into a large language model to generate a functional component call flowchart; executing the action instruction sequence corresponding to the functional component call flowchart to generate feedback information.

[0018] It should be noted that server 105 can be either hardware or software. When server 105 is hardware, it can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the server is software, it can be implemented as multiple software programs or software modules (for example, to provide feedback information generation services), or as a single software program or software module. This is not specifically limited here.

[0019] It should be noted that the feedback information generation method provided in the embodiments of the present disclosure can be executed by the server 105, or by the terminal devices 101, 102, and 103, or by the server 105 and the terminal devices 101, 102, and 103 in cooperation with each other. Accordingly, the various components (e.g., various units, subunits, modules, and submodules) included in the feedback information generation apparatus can be entirely provided in the server 105, or entirely provided in the terminal devices 101, 102, and 103, or separately provided in the server 105 and the terminal devices 101, 102, and 103.

[0020] It should be understood that Figure 1 The number of terminal devices, networks and servers in the embodiment is merely illustrative. Any number of terminal devices, networks and servers may be provided as required.

[0021] Figure 2 The process 200 of an embodiment of a method for generating feedback information is shown. The method for generating feedback information may specifically include the following steps: Step 201: Determine the application domain of the current query information and the target component set associated with the application domain.

[0022] In this embodiment, the execution subject (for example, Figure 1After obtaining the current query information, the server 105 or the terminal devices 101, 102, 103) can determine the application field of the current query information according to the semantics or source of the current query information, such as the travel field, the translation field, the financial field, etc.

[0023] Furthermore, the execution entity may determine the functional component set associated with the application field according to a preset mapping relationship table between the application field and the functional component set, and then determine the target component set associated with the application field based on the functional component set associated with the application field.

[0024] The target component set may include multiple functional components, and the functional components are used to execute action instructions to achieve corresponding functions. Different functional components correspond to different action instructions.

[0025] Here, different sets of functional components are associated with different application fields. For example, the functional components in the set of functional components associated with the travel field are mainly used to process map data; the functional components in the set of functional components associated with the translation field are mainly used to process translation data; and the functional components in the set of functional components associated with the financial field are mainly used to process financial data.

[0026] Step 202: Input the current query information and the target component set into the large language model to generate a functional component call flow chart.

[0027] In this embodiment, after obtaining the current query information and the target component set, the execution entity can input the current query information and the target component set into the large language model, and the large language model will screen the target component set according to the input current query information (that is, determine the functional components used to construct the functional component call flowchart in the target component set), process design (that is, determine the calling relationship between the screened functional components) and parameter configuration (that is, determine and configure the parameter values ​​corresponding to the parameters of the screened functional components) to generate a functional component call flowchart.

[0028] Large Language Model (LLM) refers to a type of natural language processing model trained based on deep learning technology. It has the characteristics of large parameter scale and unsupervised pre-training. There are many types of large language models, such as the GPT (Generative Pre-trained Transformer) series and the BERT (Bidirectional Encoder Representations from Transformers) series.

[0029] Here, the number of functional components used in the component call flow chart is less than the number of functional components in the target component set, that is, the large language model filters the functional components in the target component set.

[0030] Step 203: Execute the action instruction sequence corresponding to the function component call flow chart to generate feedback information.

[0031] In this embodiment, the execution subject may generate and execute a corresponding action instruction sequence according to the function component calling flow chart to obtain feedback information corresponding to the current query information.

[0032] The above-mentioned embodiment of the present disclosure determines the application field of the current query information and the target component set associated with the application field; inputs the current query information and the target component set into the large language model to generate a functional component call flowchart; executes the action instruction sequence corresponding to the functional component call flowchart to generate feedback information, thereby realizing the use of the large model to screen, design processes, and configure parameters of the target component set according to the current query information to generate a functional component call flowchart, and executes the action instruction sequence corresponding to the functional component call flowchart to generate feedback information. It fully utilizes the understanding and reasoning capabilities of the large model, does not require writing a large amount of code, and can implement the construction of the functional component call flowchart based on existing components, and does not require relying on samples for a large amount of SFT, thereby effectively improving the accuracy and efficiency of generating feedback information.

[0033] In some optional embodiments, the functional components in the functional component set are generated based on the following steps: determining the domain functions that need to be implemented in the application domain; determining multiple sub-functions based on the domain functions, and generating a functional component for each sub-function.

[0034] In this implementation, the functional components in the functional component set can be generated in the following way: the execution subject can determine the domain functions that need to be implemented according to the application field, such as travel function, translation function, etc.

[0035] Furthermore, the execution entity may split the domain function into multiple sub-functions according to one or more of the business scenario, function implementation process, and function implementation logic.

[0036] Specifically, the application field is the travel field, and the executing entity can split the field function into multiple sub-functions according to the implementation logic of the function, such as judgment sub-function, sorting sub-function, etc.

[0037] Here, the number of sub-functions depends on the degree of segmentation of the business scenario, the implementation process of the function or the implementation logic of the function. That is, the higher the degree of segmentation of the business scenario, the implementation process of the function or the implementation logic of the function, the more sub-functions are generated.

[0038] This implementation method determines the domain functions that need to be implemented in the application field; determines multiple sub-functions based on the domain functions, and generates a functional component for each sub-function, thereby achieving an effective splitting of the intelligent body functions and improving the rationality and diversity of the generated functional components.

[0039] In some optional approaches, multiple sub-functions are determined based on the domain function, including: splitting the domain function into multiple sub-functions according to different business scenarios and / or function implementation processes.

[0040] In this implementation, the execution entity can split the domain functions according to different business scenarios and / or function implementation processes.

[0041] Specifically, the application field is the travel field. The executing entity can split the domain function into multiple sub-functions according to different business scenarios, such as tourism service sub-function, travel navigation sub-function, etc.; the intelligent body function can be split into multiple sub-functions according to the implementation process of the domain function, such as search sub-function, inquiry sub-function, etc.; the domain function can also be split into multiple sub-functions according to different business scenarios and function implementation processes, such as search sub-function and inquiry sub-function in tourism service scenarios; search sub-function and inquiry sub-function in travel navigation scenarios, etc. This application does not limit this.

[0042] This implementation method realizes the splitting of domain functions based on business scenarios and / or function implementation processes, further improving the rationality and effectiveness of the split sub-functions.

[0043] In some optional embodiments, the application area includes the travel area.

[0044] In this implementation, if the executing entity determines that the application field corresponding to the current query information is the travel field based on the mapping relationship between the preset query information type and the application field, the target component set associated with the travel field and the current query information can be input into the large language model to generate a component call flowchart, and the action instruction sequence corresponding to the component call flowchart is executed to generate feedback information.

[0045] Here, the target component set associated with the travel field may include multiple functional components for processing map data, such as a search component, a query component, and the like.

[0046] The search component is used to search for location locations and / or routes between locations; the query component is used to query for associated information of location locations and / or associated information of routes.

[0047] Specifically, the current query information is "Where is place A?", and the application field of the current query information is the travel field. The execution entity can input the target component set related to the travel field (for example, search components and query components) and the current query information into the large language model, generate a functional component call flowchart, and execute the action instruction sequence corresponding to the functional component call flowchart to generate feedback information.

[0048] This implementation method realizes the generation of feedback information in the travel field by setting the application field to include the travel field.

[0049] In some optional embodiments, the functional component set includes at least one of the following: a point search component, a general route search component, a sub-route search component, a general route-asking component, a sub-route-asking component, a point-asking component, a clarification component, and a prompt component.

[0050] In this implementation, if the functional components in the functional component set are generated in the following manner: the domain function is split into multiple sub-functions according to the implementation process of the domain function, and a functional component is generated for each sub-function, then the functional component set may include at least one of the following: a point search component, a general route search component, a sub-route search component, a general route inquiry component, a sub-route inquiry component, a point inquiry component, a clarification component, and a prompt component.

[0051] Among them, the point search component can be used to search for location; the general route search component can be used to search for routes between locations; the sub-route search component can be used to search for routes between locations corresponding to transportation methods; the general route-asking component can be used to inquire about the associated information of routes; the sub-route-asking component can be used to inquire about the associated information corresponding to routes and transportation methods; the point-asking component can be used to inquire about the associated information of location; the clarification component can be used to retrieve sample query information; and the prompt component can be used to output prompt information if the sample query information processing fails.

[0052] Here, the number of sub-route search components may include multiple, and different sub-route search components correspond to different modes of transportation. For example, the number of sub-route search components may be four, namely the first sub-route search component, the second sub-route search component, the third sub-route search component and the fourth sub-route search component. The transportation mode corresponding to the first sub-route search component is walking, the transportation mode corresponding to the second sub-route search component is cycling, the transportation mode corresponding to the third sub-route search component is public transportation, and the transportation mode corresponding to the fourth sub-route search component is driving.

[0053] There may be multiple sub-components for directions, and different sub-components for directions correspond to different modes of transportation. For example, there may be four sub-components for directions, namely, a first sub-component for directions, a second sub-component for directions, a third sub-component for directions, and a fourth sub-component for directions. The first sub-component for directions corresponds to walking, the second sub-component for directions corresponds to cycling, the third sub-component for directions corresponds to public transportation, and the fourth sub-component for directions corresponds to driving.

[0054] Specifically, the current query information is "where is the best sweet shrimp in place A", the application field is the travel field, and the set of functional components associated with the application field may include a point search component, a general route search component, a sub-route search component, a general route-asking component, and a sub-route-asking component.

[0055] This implementation method sets a functional component set including at least one of the following: a point search component, a general route search component, a sub-route search component, a general route inquiry component, a sub-route inquiry component, a point inquiry component, a clarification component, and a prompt component through an implementation process based on the domain functions of the travel field. This realizes the effective configuration of the functional component set, improves the richness and comprehensiveness of the determined target component set, and thereby improves the effectiveness of the determined functional component call flowchart.

[0056] In some optional methods, the parameters of the point search component include at least one of a place name parameter, a place category parameter, a place feature parameter, and a place quantity parameter; the parameters of the total route search component include at least one of a transportation mode parameter and a time point parameter; the parameters of the sub-route search component include a time point parameter; the parameters of the total route-asking component and the sub-route-asking component respectively include a route intention parameter; the parameters of the point-asking component include a place intention parameter; the parameters of the clarification component include at least one of a clarification content parameter and a query intention parameter; and the parameters of the prompt component include a complexity parameter.

[0057] In this implementation, the functional component set may include at least one of a point search component, a general route search component, a sub-route search component, a general route-asking component, a sub-route-asking component, a point-asking component, a clarification component, and a prompt component.

[0058] The parameters of the point search component may include at least one of a place name parameter, a place category parameter, a place feature parameter, and a place quantity parameter.

[0059] The parameters of the overall route search component may include at least one of a traffic mode parameter and a time point parameter, the parameters of the sub-route search component may include a time point parameter, and the parameters of the overall route asking component and the sub-route asking component may respectively include a route intention parameter.

[0060] Specifically, the current query information is "How long does it take to take the subway from place A to place B at 6 pm on Sunday?" The functional components used in the functional component call flowchart may include a point search component, a total route search component, and a total direction-asking component. The calling relationship between the functional components is that the total direction-asking component calls the total route search component, and the total route search component calls the point search component. The large language model determines based on the current query information that the parameter values ​​that match the place name parameters of the point search component are place A and place B, the parameter values ​​that match the transportation mode parameters and time point parameters of the total route search component are subway and 6 pm on Sunday respectively, and the parameter value that matches the route intention parameters of the total direction-asking component is the overall time.

[0061] The question point component may include a location intent parameter.

[0062] Specifically, the current query information is "Where is the best place to eat sweet shrimp in place A?" The functional components used in the functional component call flow chart may include two search point components (a first search point component and a second search point component) and a question point component. The calling relationship between the functional components is that the second search point component calls the first search point component, and the question point component calls the second search point component. The parameters of the first search point component may include a place name parameter, and the parameters of the second search point component may include a place category parameter, a place feature parameter, and a place quantity parameter. The large language model determines based on the current query information that the parameter value that matches the place name parameter is place A, the parameter value that matches the place category parameter is delicious food, the parameter value that matches the place feature parameter is sweet shrimp, the parameter value that matches the place quantity parameter is n, and the parameter value that matches the place intent parameter of the question point component is a search introduction.

[0063] The clarification component may include at least one of a clarification content parameter and a query intent parameter, and the prompt component includes a complexity parameter.

[0064] Specifically, the current query information is "Provide a recommended route?", the functional components used in the functional component call flowchart may include a clarification component, the parameters of the clarification component may include a clarification content parameter, and the parameter value that matches the clarification content parameter determined by the large language model based on the current query information may be what the starting point and end point are; the current query information is "Please plan the best route from A to B for me based on real-time traffic conditions and historical data. The route needs to consider the following conditions: avoid all congested sections, avoid construction areas, give priority to public transportation, and provide detailed information and estimated arrival time for each transfer point", the functional components used in the functional component call flowchart may include a prompt component, the parameters of the prompt component may include a complexity parameter, and the parameter value that matches the complexity parameter determined by the large language model based on the current query information may be that the calculation steps exceed 10 steps.

[0065] This implementation method realizes flexible configuration of target components by setting the parameters of the point search component to include at least one of a place name parameter, a place category parameter, a place feature parameter, and a place quantity parameter; the parameters of the overall route search component to include at least one of a transportation mode parameter and a time point parameter; the parameters of the sub-route search component to include a time point parameter; the parameters of the overall route-asking component and the sub-route-asking component to include a route intention parameter; the parameters of the point-asking component to include a place intention parameter; the parameters of the clarification component to include at least one of a clarification content parameter and a query intention parameter; and the parameters of the prompt component to include a complexity parameter.

[0066] In some optional methods, the current query information and the target component set are input into the large language model to generate a functional component call flowchart, including: arranging prompt words according to the current query information and the target component set to generate an arrangement result; inputting the arrangement result into the large language model to generate a functional component call flowchart.

[0067] In this implementation, the execution subject can arrange prompt words according to the current query information and the target component set to generate an arrangement result.

[0068] Here, the orchestration results may include task instructions, output requirements, etc., where task instructions are used to indicate the required components and the goals to be completed, such as query, calculation, generation, etc.; output requirements are used to specify the format of the results (such as table, JSON (JavaScript Object Notation), natural language, etc.) and content granularity (such as whether comparative analysis is required, etc.).

[0069] Furthermore, the execution entity can input the arrangement results into the large language model to generate a functional component call flow chart.

[0070] This method arranges prompt words based on the current query information and the target component set to generate an arrangement result; the arrangement result is input into the large language model to generate a functional component call flowchart. That is, the large language model is guided by the arrangement result to generate a component call flowchart, further improving the accuracy and reliability of the generated component call flowchart.

[0071] Further references Figure 3 , which shows Figure 2 FIG. 3 is a flow chart of another embodiment of a method for generating feedback information. In this embodiment, the flow chart 300 of the method for generating feedback information may include the following steps: Step 301: Determine the application field of the current query information and the set of functional components associated with the application field.

[0072] In this embodiment, after determining the application field of the current query information, the execution subject may determine the functional component set associated with the application field according to a preset mapping relationship table between the application field and the functional component set.

[0073] Step 302: Screen the functional component set based on the current query information to generate a target component set.

[0074] In this embodiment, the manner in which the execution subject filters the functional component set based on the current query information may be associated with the manner in which the functional components in the functional component set are constructed.

[0075] Among them, if the functional components in the functional component set are constructed based on business scenarios (for example, tourism service scenarios, travel navigation scenarios, etc.), the execution entity can first perform semantic recognition on the current query information, determine the business scenario corresponding to the current query information, and then filter the functional component set according to the determined business scenario to obtain the target component set.

[0076] Specifically, if the functional component set includes a component subset corresponding to the tourism service scenario and a component subset corresponding to the travel navigation scenario, and the business scenario corresponding to the current query information is the tourism service scenario, the component subset corresponding to the tourism service scenario can be determined as the target component set.

[0077] If the functional components in the functional component set are constructed based on the implementation process of the domain function (for example, if the implementation process of the domain function is inquiry, the functional component set may include components corresponding to the inquiry; if the implementation process of the domain function is search-inquiry, the functional component set may include components corresponding to the inquiry and components corresponding to the search, wherein the search can be further divided into search points, search routes, etc., and the inquiry can be further divided into question points, question routes, etc., that is, the component subset corresponding to the inquiry can include components corresponding to the question points and components corresponding to the question routes, and the component subset corresponding to the search can include components corresponding to the search points and components corresponding to the search routes), the execution subject can first perform semantic recognition on the current query information, determine the process required to answer the current query information, and filter the functional component set based on the required process to obtain the target component set.

[0078] Specifically, if the functional component set includes a component subset corresponding to the query and a component subset corresponding to the search, and the process required to answer the current query information is a query, the component subset corresponding to the query can be determined as the target component set.

[0079] If the functional components in the functional component set are constructed based on the implementation process of business scenarios and domain functions, the execution entity first performs semantic recognition on the current query information, determines the business scenario corresponding to the current query information and the process required to answer the current query information, and then preliminarily screens the functional component set based on the business scenario. Based on the process required to answer the current query information, the preliminarily screened functional component set is further screened to obtain the target component set.

[0080] Specifically, if the functional component set includes a component subset corresponding to the tourism service scenario and a component subset corresponding to the travel navigation scenario, and the component subset corresponding to the tourism service scenario and the component subset corresponding to the travel navigation scenario respectively include a component subset corresponding to the inquiry and a component subset corresponding to the search; and the business scenario corresponding to the current query information is the tourism service scenario, and the process required to reply to the current query information is an inquiry, then the component subset corresponding to the inquiry process in the component subset corresponding to the tourism service scenario can be determined as the target component set.

[0081] Step 303: Input the current query information and the target component set into the large language model to generate a functional component call flow chart.

[0082] In this embodiment, the implementation details and technical effects of step 303 can be found in the description of step 202 and will not be repeated here.

[0083] Step 304: Execute the action instruction sequence corresponding to the function component call flow chart to generate feedback information.

[0084] In this embodiment, the implementation details and technical effects of step 304 can be found in the description of step 203 and will not be repeated here.

[0085] The above-mentioned embodiments of the present disclosure determine the application field of the current query information and the set of functional components associated with the application field; filter the set of functional components based on the current query information to generate a target component set; input the current query information and the target component set into a large language model to generate a functional component call flowchart; execute the action instruction sequence corresponding to the functional component call flowchart to generate feedback information, thereby realizing the filtering of the functional component set input into the large language model, narrowing the filtering range of the large language model, and further improving the efficiency and accuracy of generating feedback information.

[0086] In some optional manners, filtering the functional component set based on the current query information to generate the target component set includes: filtering the functional component set based on the current query information and associated information of the current query information to generate the target component set.

[0087] In this implementation, the execution subject may fuse the current query information and the associated information of the current query information to obtain fused information, and screen the functional component set based on the fused information to generate a target component set.

[0088] The associated information of the current query information includes at least one of historical query information, user portrait information and current environment information.

[0089] Here, the way of screening the functional component set may be associated with the way of constructing the functional components in the functional component set.

[0090] Specifically, if the functional components in the functional component set are built based on business scenarios, the execution entity can first determine the corresponding business scenario based on the current query information, at least one of the historical query information, user portrait information and current environment information, and then filter the functional component set according to the determined business scenario to obtain the target component set.

[0091] If the functional components in the functional component set are constructed based on the implementation process of the domain functions, the execution entity may first determine the process required to answer the current query information based on at least one of the historical query information, user portrait information and current environment information, and the current query information, and filter the functional component set based on the required process to obtain the target component set.

[0092] If the functional components in the functional component set are constructed based on the implementation process of business scenarios and domain functions, the execution entity may first determine the corresponding business scenario and the process required to answer the current query information based on at least one of the historical query information, user portrait information and current environment information, and the current query information, and then preliminarily screen the functional component set based on the business scenario, and further screen the preliminarily screened functional component set based on the process required to answer the current query information to obtain the target component set.

[0093] Here, the historical query information may include one or more query information submitted by the user before the current query information, which has a semantic association with the current query information or does not have a semantic association with the current query information.

[0094] User portrait information may include multiple items, such as demographic information (such as gender, age, etc.), user search or query records (such as locations, routes, points of interest, etc.), search frequency and time period (such as querying commuting routes during rush hours in the morning and evening, searching for leisure places on weekends, etc.), common transportation methods (such as driving, bus, subway, etc.), travel time patterns (such as weekday commuting time, weekend travel time), social relationships (such as location sharing records with friends, interest tags synchronized on social platforms for common travel routes, etc.), related consumption data (such as interaction with e-commerce platforms, interaction with life service accounts, etc.), life service preferences (consumption habits, dining preferences, travel scenarios, etc.), interactive behaviors (such as place collection, comments or ratings on points of interest, sharing operations, etc.), etc.

[0095] The current environmental information may include multiple items, such as topographic data (such as altitude, vegetation cover, etc.), road and traffic networks (such as the number of lanes, traffic rules, etc.), point of interest information (such as static points of interest, dynamic points of interest, etc.), traffic status (real-time road conditions, parking lot information, etc.), weather and climate (such as current weather, warning information, etc.), human flow and activities (such as crowd density, large-scale events, etc.), device sensor data (such as GPS (Global Positioning System) signal strength, gyroscope / accelerometer data, etc.), culture and customs (such as religion, ethnic minority settlements, etc.), etc.

[0096] Specifically, the functional components in the functional component set are constructed based on the implementation process of business scenarios and domain functions. If the current query information is "How long does it take to get from A to B?", the user portrait information includes: commonly used transportation methods, such as driving; the current environmental information includes: current weather, such as heavy rain; the execution entity can fuse the current query information, user portrait information and current environmental information to generate fused information, and based on the fused information, determine the business scenario corresponding to the current query information and the process required to respond to the current query information.

[0097] Here, before fusing the current query information and the associated information of the current query information, the execution entity may pre-process the associated information of the current query information, and fuse the current query information and the pre-processed associated information to generate fused information.

[0098] Among them, the preprocessing methods may include multiple methods. For example, if the related information of the current query information includes historical query information, the historical query information can be sorted by time and trend features can be extracted, such as fluctuations in question frequency, topic evolution, etc.; if the related information of the current query information includes current environmental information, it can be identified whether the current query information is triggered by an emergency environmental event. If so, the current environmental information can be filtered according to the emergency environmental event to generate filtered current environmental information; if the related information of the current query information includes user portrait information, the user portrait information can be filtered according to the current query to generate filtered user portrait information.

[0099] It should be pointed out that if the associated information of the current query information includes historical query information and user portrait information, if the number of historical query information is less than the first preset value, for example, for a new user, the trend feature extraction fails, then the trend features in the group characteristics of similar users of the current user can be determined based on the user portrait information, and the trend features of the current user can be filled in based on the determined trend features.

[0100] Furthermore, if the associated information of the current query information includes historical query information, user portrait information and current environment information, the execution entity can fuse the preprocessed historical query information, the first weight of the historical query information, the preprocessed user portrait information, the second weight of the user portrait information, the preprocessed current environment information, the third weight of the current environment information, and the current query information to generate fused information.

[0101] Here, there are multiple ways to set the first weight, the second weight and the third weight. For example, if the current query information is triggered by an emergency environmental event, the third weight can be set to be greater than the second weight and the first weight, that is, the third weight is the largest; if the current query information is not triggered by an emergency environmental event, and the number of historical query information is less than the preset value, the second weight can be set to be greater than the first weight and the third weight, that is, the second weight is the largest; if the current query information is not triggered by an emergency environmental event, and the number of historical query information is greater than the second preset value (the second preset value is greater than the first preset value), such as an old user, the first weight can be set to be greater than the second weight and the third weight, that is, the first weight is the largest.

[0102] This implementation method generates a target component set by screening a set of functional components based on the current query information and the associated information of the current query information, thereby further reducing the number of functional components input into the large language model, narrowing the screening range of the large language model, and further improving the efficiency and accuracy of determining feedback information.

[0103] In some optional methods, the current query information and the target component set are input into the large language model to generate a functional component call flowchart, including: inputting the current query information, the associated information of the current query information and the target component set into the large language model to generate a functional component call flowchart.

[0104] In this implementation, the execution entity may directly input the historical query information, user portrait information, at least one item of the current environment information, the current query information, and the target component set into the large language model to generate a functional component call flowchart; or it may first arrange the prompt words based on the historical query information, user portrait information, at least one item of the current environment information, the current query information, and the target component set to generate an arrangement result, and then input the arrangement result into the large language model to generate a functional component call flowchart. This application does not limit this.

[0105] Here, after inputting historical query information, at least one item of user portrait information and current environment information, current query information and target component set into the large language model, the large language model can screen the functional components in the target component set based on the historical query information, at least one item of user portrait information and current environment information, and current query information, determine the functional components required to construct the functional component call flowchart and the calling relationship between the functional components, and determine the parameter values ​​corresponding to the parameters of the functional components required to construct the functional component call flowchart based on the above-mentioned historical query information, at least one item of user portrait information and current environment information, and current query information, and configure the functional components according to the parameter values.

[0106] This implementation method generates a functional component call flowchart by inputting the current query information, the associated information of the current query information and the target component set into a large language model. It takes into account the impact of the associated information of the current query information on the generated functional component call flowchart, thereby improving the accuracy of the generated functional component call flowchart.

[0107] Continue to see Figure 4 , Figure 4 It is a schematic diagram of an application scenario of the feedback information generating method according to this embodiment.

[0108] The current query information 401 is "How long does it take to get from point A to point B at 6 pm on Sunday?" The execution entity 402 can determine the application field of the current query information 401 (e.g., the travel field) and a set of functional components associated with the application field. The functional components are used to execute action instructions. Furthermore, based on at least one item 403 of historical query information, user profile information, and current environmental information, as well as the current query information, the set of functional components is filtered to obtain a target component set 404. For example, the target component set includes: a point search component, a general route search component, a sub-route search component, a general route inquiry component, a sub-route inquiry component, and a clarification component.

[0109] Furthermore, the execution entity may input at least one of the historical query information, user profile information, and current environment information 403, the current query information 401, and the target component set 404 into the large language model 405 to generate a functional component call flowchart 406, wherein the number of functional components (for example, a point search component, a total route search component, and a total route-asking component) used in the component call flowchart 406 is less than the number of functional components in the target component set.

[0110] Furthermore, the execution subject may execute the action instruction sequence 407 corresponding to the function component call flow chart to generate feedback information 408. Here, the action instruction sequence may be as follows: 1. p1 = search_for_ poi (name = 'Place A') p2 = search_for_ poi (name = 'B') / / Search point components to find the first occurrence of A and B 2.r1 = search_for_rsn (transportation mode = 'subway', departure time = 'Sunday 06:00pm', starting point = p1, end point = p2) / / Total route search component, departure time is [Sunday 06:00pm], transportation mode is [subway], starting point is [p1], end point is [p2], check whether there is a standard route 3. m1 = ask_for_navigation (target = 'overall time', route_target = [r1]) / / The total route-asking component queries [r1] for information about [overall time] Further references Figure 5 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of a feedback information generating device. Figure 2 The method embodiment shown corresponds to the embodiment shown.

[0111] like Figure 5 As shown, the feedback information generating device 500 of this embodiment includes: a determination module 501 , an input module 502 and an execution module 503 .

[0112] The determination module 501 may be configured to determine the application field of the current query information and a target component set associated with the application field.

[0113] The input module 502 may be configured to input current query information and a target component set into the large language model to generate a functional component call flow chart.

[0114] The generation module 503 may be configured to execute the action instruction sequence corresponding to the functional component call flow chart to generate feedback information.

[0115] In some optional embodiments of this embodiment, the determination module includes: a determination unit and a screening unit, wherein the determination unit can be configured to determine the application field of the current query information and the set of functional components associated with the application field; the screening unit can be configured to screen the set of functional components based on the current query information to generate a target component set.

[0116] In some optional aspects of this embodiment, the screening unit may be further configured to screen the functional component set based on the current query information and associated information of the current query information to generate a target component set.

[0117] In some optional aspects of this embodiment, the input module may be further configured to: input the current query information, associated information of the current query information, and a target component set into the large language model to generate a functional component call flow chart.

[0118] In some optional embodiments of this embodiment, the functional components in the functional component set are generated based on the following steps: determining the domain functions that need to be implemented in the application domain; determining multiple sub-functions based on the domain functions, and generating a functional component for each sub-function.

[0119] In some optional aspects of this embodiment, multiple sub-functions are determined based on the domain function, including: splitting the domain function into multiple sub-functions according to different business scenarios and / or function implementation processes.

[0120] In some optional aspects of this embodiment, the application field includes the travel field.

[0121] In some optional embodiments of this embodiment, the functional component set includes at least one of the following: a point search component for searching for location locations; a general route search component for searching for routes between locations; a sub-route search component for searching for routes between locations corresponding to transportation modes; a general route inquiry component for inquiring about associated information of routes; a sub-route inquiry component for inquiring about associated information corresponding to routes and transportation modes; a point inquiry component for inquiring about associated information of location locations; a clarification component for reacquiring the sample query information; and a prompt component for outputting prompt information of failure in processing the sample query information.

[0122] In some optional embodiments of this embodiment, the parameters of the point search component include at least one of a place name parameter, a place category parameter, a place feature parameter, and a place quantity parameter; the parameters of the overall route search component include at least one of a transportation mode parameter and a time point parameter; the parameters of the sub-route search component include a time point parameter; the parameters of the overall route-asking component and the sub-route-asking component respectively include a route intention parameter; the parameters of the point-asking component include a place intention parameter; the parameters of the clarification component include at least one of a clarification content parameter and a query intention parameter; and the parameters of the prompt component include a complexity parameter.

[0123] In some optional embodiments of this embodiment, the input module further includes: an arrangement unit and an input unit. The arrangement unit can be configured to arrange prompt words based on the current query information and the target component set to generate an arrangement result; the input unit can be configured to input the arrangement result into the large language model to generate a functional component call flowchart.

[0124] In the technical solutions disclosed herein, the acquisition, storage, and application of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0125] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0126] Figure 6 A schematic block diagram of an example electronic device 600 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0127] like Figure 6 As shown, device 600 includes a computing unit 601, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 602 or a computer program loaded from a storage unit 608 into a random access memory (RAM) 603. RAM 603 may also store various programs and data required for the operation of device 600. Computing unit 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to bus 604.

[0128] Various components in device 600 are connected to I / O interface 605, including an input unit 606, such as a keyboard, mouse, etc.; an output unit 607, such as various types of displays, speakers, etc.; a storage unit 608, such as a magnetic disk, optical disk, etc.; and a communication unit 609, such as a network card, modem, wireless communication transceiver, etc. The communication unit 609 allows device 600 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0129] The computing unit 601 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 601 performs the various methods and processes described above, such as the method for generating feedback information. For example, in some embodiments, the method for generating feedback information can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the method for generating feedback information described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the method for generating feedback information through any other suitable means (e.g., via firmware).

[0130] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0131] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0132] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0133] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0134] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0135] A computer system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a host product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosts and virtual private server (VPS) services. Servers can also be classified as distributed system servers or servers integrated with blockchain.

[0136] According to the technical solution of the embodiment of the present disclosure, the accuracy and efficiency of generating feedback information are effectively improved.

[0137] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions provided by this disclosure can be achieved. This is not limited herein.

[0138] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for generating feedback information, comprising: Determine an application field of the current query information and a target component set associated with the application field, wherein the target component set includes a plurality of functional components, and the functional components are used to execute action instructions; Inputting the current query information and the target component set into a large language model to generate a function component call flow chart, wherein the number of function components used in the component call flow chart is less than the number of function components in the target component set; Execute the action instruction sequence corresponding to the functional component call flow chart to generate feedback information.

2. The method according to claim 1, wherein The determining of the application field of the current query information and the target component set associated with the application field includes: Determine the application domain of the current query information and a set of functional components associated with the application domain; The functional component set is screened based on the current query information to generate a target component set.

3. The method according to claim 2, wherein: The screening of the functional component set based on the current query information to generate a target component set includes: Based on the current query information and the associated information of the current query information, the functional component set is screened to generate a target component set, wherein the associated information of the current query information includes at least one of historical query information, user portrait information and current environment information.

4. The method according to claim 3, wherein: The step of inputting the current query information and the target component set into a large language model to generate a functional component call flow chart includes: The current query information, the associated information of the current query information and the target component set are input into a large language model to generate a functional component call flow chart.

5. The method according to claim 2, wherein: The functional components in the functional component set are generated based on the following steps: Determine the domain functions that need to be implemented in the application field; A plurality of sub-functions are determined based on the domain function, and a functional component is generated for each sub-function.

6. The method according to claim 5, wherein: The determining of multiple sub-functions based on the domain function includes: According to different business scenarios and / or function implementation processes, the domain functions are split into multiple sub-functions.

7. The method according to claim 2, wherein: The application fields include the travel field.

8. The method according to claim 7, wherein: The functional component set includes at least one of the following: Search point component, used to search for location; The overall route search component is used to search for routes between locations; Sub-route search component, used to search for routes between locations corresponding to transportation methods; The general route-asking component is used to inquire about route-related information; The sub-direction component is used to inquire about the associated information between routes and transportation methods; The query point component is used to query the related information of the location; a clarification component, configured to retrieve the sample query information; And, a prompt component is used to output prompt information when the sample query information processing fails.

9. The method according to claim 8, wherein The parameters of the point search component include at least one of a place name parameter, a place category parameter, a place feature parameter, and a place quantity parameter. The parameters of the overall route search component include at least one of a traffic mode parameter and a time point parameter, and the parameters of the sub-route search component include a time point parameter. The parameters of the main route-asking component and the sub-route-asking component respectively include route intention parameters, The parameters of the question point component include location intention parameters, The parameters of the clarification component include at least one of a clarification content parameter and a query intent parameter. The parameters of the prompt component include a complexity parameter.

10. The method according to any one of claims 1 to 9, wherein: Inputting the current query information and the target component set into a large language model to generate a functional component call flow chart, including: Arrange prompt words according to the current query information and the target component set, and generate an arrangement result; The arrangement result is input into a large language model to generate a functional component call flow chart.

11. A feedback information generating device, comprising: a determination module configured to determine an application field of the current query information and a target component set associated with the application field, wherein the target component set includes a plurality of functional components, and the functional components are used to execute action instructions; an input module configured to input the current query information and the target component set into a large language model to generate a function component call flow chart, wherein the number of function components used in the component call flow chart is less than the number of function components in the target component set; The execution module is configured to execute the action instruction sequence corresponding to the functional component call flow chart and generate feedback information.

12. An electronic device, characterized in that: include: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 10.

13. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 10.

14. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 10.

Images